Analog engineering practice education method, device, equipment and medium

Abstract

The application relates to the technical field of data processing, and in particular to a simulation engineering practice education method, device, equipment and medium. The method comprises the following steps: providing personalized experience for students through identity recognition, loading a multi-agent role model by using a role simulation instruction, helping students understand engineering cooperation from different role perspectives, generating role answers and initial product demand documents based on user needs, cultivating demand analysis and management capabilities, receiving user decisions in real time and generating decision effects, enhancing decision practice feedback, improving iterative optimization and critical thinking through correction documents and independent evaluation of decisions, and finally visualizing the correction documents and evaluation results, promoting knowledge integration and reflection, forming a complete practice closed loop, and effectively improving the comprehensive ability of students to solve complex engineering problems.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, equipment and medium for simulating engineering practice education. Background Technology

[0002] Engineering education currently faces significant challenges, particularly in cultivating students' practical engineering skills. According to reports, over 70% of engineering graduates nationwide find themselves lacking hands-on experience and problem-solving abilities after entering the workforce. This issue not only impacts graduates' employment competitiveness but also hinders innovation and development in engineering education at universities.

[0003] To address these challenges, the education sector has explored various methods to improve engineering practice education. For example, offline project training involves students in real engineering projects to enhance their practical skills, while role-playing teaching methods use scripts and assigned roles to allow students to complete tasks in simulated scenarios. While these methods have alleviated resource shortages to some extent, they still have many limitations.

[0004] Existing technological approaches still have significant shortcomings in addressing the shortage of resources for engineering practice education. First, offline project training requires substantial human and material resources, making it difficult to meet the demands of large-scale teaching. Second, role-playing teaching methods lack personalized support, failing to fully mobilize students' initiative and creativity. These issues prevent students from gaining opportunities to practice under the guidance of high-level architects, product managers, project managers, and other key roles, while also limiting their opportunities to interact with experts from multiple fields, thus impacting the development of their comprehensive abilities. These deficiencies severely restrict the cultivation of students' engineering practice skills and make it difficult to meet the comprehensive engineering competence requirements of engineering education accreditation. Summary of the Invention

[0005] To enhance students' comprehensive ability to solve complex engineering problems, this application provides a simulated engineering practice education method, device, equipment, and medium.

[0006] Firstly, this application provides a simulated engineering practice education method, which adopts the following technical solution: A simulated engineering practice education method includes: Obtain the current user's identity, which is either a student or a teacher; When the user is a student, the system receives the role simulation instruction corresponding to the current user and loads the target role model corresponding to the role simulation instruction. The role simulation instruction is used to trigger the current user to select multiple intelligent agents to be invoked, with one intelligent agent corresponding to one role model. Receive the user requirements input by the current user, and based on the user requirements, generate the role response corresponding to each target role model, and generate an initial product requirement document based on the user requirements and the role response; The system receives user decisions input by the current user in real time and generates the corresponding decision effects. Based on the user's decision and the effect of the decision, the initial product requirement document is revised to obtain the revised product requirement document. Based on the effects of each decision and the revised product requirements document, each user's decision is independently evaluated to obtain an independent evaluation result, and the revised product requirements document and the independent evaluation result are displayed.

[0007] By adopting the above technical solutions, personalized experiences are provided to students through identity recognition; multi-agent role models are loaded using role simulation instructions to help students understand engineering collaboration from different role perspectives; role responses and initial product requirement documents are generated based on user needs, cultivating requirements analysis and management capabilities; user decisions are received in real time and decision effects are generated, enhancing feedback on decision-making practice; iterative optimization and critical thinking are improved through document revision and independent evaluation of decisions; and finally, the revised documents and evaluation results are visualized to promote knowledge integration and reflection, forming a complete practical closed loop and effectively improving students' comprehensive ability to solve complex engineering problems.

[0008] In one possible implementation, the step of receiving the user decision input by the current user in real time and generating the decision effect corresponding to the user decision includes: Receive the user decision of the current user; Based on the user needs and user decisions, extract the key decision parameters from each user decision; Based on the pre-built engineering impact model, analyze the impact parameters of each user decision on subsequent stages; Based on the key decision parameters and the influencing parameters, at least two virtual project progress paths are generated, each corresponding to the engineering results under different decision combinations. Problem decision nodes are marked based on the engineering results, and decision effects are generated.

[0009] In one possible implementation, after obtaining the revised product requirements document, the method further includes: Receive the document generation instruction from the current user and determine the completeness of the revised product requirement document; Based on the completeness, determine whether a final product requirements document can be generated; When it is determined that the final product requirement document can be generated, the final product requirement document corresponding to the current user is generated.

[0010] In one possible implementation, after obtaining independent evaluation results for each user's decision, the method further includes: Based on the user needs and the user decisions, evidence of the current user's contribution to the product requirement document generation process is collected. Call upon engineering training expert agents to generate agent scores and improvement suggestions based on multiple engineering capability dimensions; Receive teacher ratings, correct the agent ratings based on the teacher ratings to obtain a comprehensive rating, and output the comprehensive rating.

[0011] In one possible implementation, based on the user needs, a role response corresponding to each target role model is generated, including: Identify key information in the user's needs, including at least one of product information, action information, and need information; Based on the aforementioned key information, the scope of responsibilities and interaction guidance degree corresponding to each target role model are determined. The interaction guidance degree is a parameter representing the ratio between the agent's guiding behavior and direct output behavior during the dialogue process. Based on the interaction guidance degree, scope of responsibility, and key information corresponding to each target role model, the corresponding intelligent agent domain knowledge base is triggered to generate a professional response to obtain the role answer corresponding to each target role model.

[0012] In one possible implementation, the method further includes: Obtain the specific functional tasks corresponding to each intelligent agent; Based on the user's needs and the specific functional tasks, a recommended intelligent agent is determined.

[0013] In one possible implementation, after obtaining the current user's identity, the method further includes: When the user is a teacher, obtain the user instruction corresponding to the current user. The user instruction is used for one of the following: issuing training tasks, setting evaluation criteria, and manual score correction. When the user instruction is a manual rating correction, the training information corresponding to the current user is obtained and displayed. The training information includes the input content of each student, the called agent, and the evaluation result. Receive the student rating input by the current user and add the student rating to the corresponding student's teacher rating.

[0014] Secondly, this application provides a simulated engineering practice education device, which adopts the following technical solution: A simulated engineering practice education device, comprising: The acquisition module is used to acquire the user identity of the current user, which is either a student or a teacher; The loading module is used to receive the role simulation instruction corresponding to the current user when the user's identity is a student, and load the target role model corresponding to the role simulation instruction. The role simulation instruction is used to trigger the current user to select multiple intelligent agents to be invoked, and one intelligent agent corresponds to one role model. The receiving module is used to receive the user requirements input by the current user, generate a role response corresponding to each target role model based on the user requirements, and generate an initial product requirement document based on the user requirements and the role responses. The generation module is used to receive the user decision input by the current user in real time and generate the decision effect corresponding to the user decision; The modification module is used to modify the initial product requirement document based on the user's decision and the effect of the decision, so as to obtain the modified product requirement document. The display module is used to independently evaluate each user's decision based on the effect of each decision and the revised product requirement document, obtain independent evaluation results, and display the revised product requirement document and the independent evaluation results.

[0015] Thirdly, this application provides an electronic device that adopts the following technical solution: An electronic device comprising: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: perform the method described in any of the first aspects above.

[0016] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium includes: a computer program stored thereon that can be loaded by a processor and execute the method described in any of the first aspects above.

[0017] In summary, this application includes the following beneficial technical effects: By providing personalized experiences for students through identity recognition, and loading multi-agent role models using role simulation instructions, students can understand engineering collaboration from different role perspectives. Based on user needs, role responses and initial product requirement documents are generated, cultivating requirements analysis and management capabilities. User decisions are received in real time and decision effects are generated, enhancing feedback on decision-making practice. By revising documents and independently evaluating decisions, iterative optimization and critical thinking are improved. Finally, the revised documents and evaluation results are visualized, promoting knowledge integration and reflection, forming a complete practical closed loop, and effectively improving students' comprehensive ability to solve complex engineering problems. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a simulated engineering practice education method provided in an embodiment of this application; Figure 2 This is a block diagram of a simulated engineering practice education device provided in an embodiment of this application; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] The following is in conjunction with the appendix Figure 1 - Appendix Figure 3 This application will be described in further detail.

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] To facilitate understanding of the technical solutions proposed in this application, several elements that will be introduced in the description of this application are first presented here. It should be understood that the following description is only for the purpose of understanding these elements and the content of the embodiments of this application, and does not necessarily cover all possible situations.

[0022] To overcome these bottlenecks, the MAESTRO Multi-Agent Engineering Simulation and Training for Real-world Operations model was proposed. Specifically, this model simulates complex engineering roles and processes through multi-agent collaboration, enabling students to engage in engineering role-playing and interaction. Students gain comprehensive engineering knowledge and skills training on a single platform that requires no human supervision, making it affordable and accessible to every student. This provides a novel solution for engineering skills training and evaluation, offering a completely new approach. This application provides a simulated engineering practice education method, such as... Figure 1 As shown, the method provided in this embodiment is executed by an electronic device equipped with a simulated engineering practice education platform. The simulated engineering practice education method is applied to this platform. The electronic device can be a server or a terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication. This embodiment does not impose any limitations on this connection. The method includes steps S101-S106, wherein: Step S101: Obtain the current user's identity.

[0023] The user can be identified as a student or a teacher.

[0024] Specifically, when a user enters the simulated engineering practice education platform, they typically input their username and password. Upon receiving this input, the electronic device identifies the user based on the username, verifies the correctness of the username and password, and then allows the user to access the platform. More specifically, the username can be compared with accounts and corresponding identities in a pre-set database to determine the current user's identity. The simulated engineering practice education platform, installed on the electronic device, provides students with a simulated platform to help them understand engineering collaboration from different role perspectives.

[0025] More specifically, the simulated engineering practice education platform provides multiple trained agents. After entering the platform, students can choose an agent and interact with it, experiencing the entire process from requirements analysis, design and development, testing and deployment to delivery. During this process, students can switch between multiple interdisciplinary and cross-domain agents, gaining a deeper understanding of the logic and interrelationships of each engineering stage from different perspectives. The agents are trained using a combination of large models and domain knowledge bases, enabling them to independently complete tasks within their scope of responsibility. For example, a project manager agent can display project schedules, while a test engineer agent can analyze code performance and generate test reports.

[0026] Step S102: When the user's identity is a student, receive the role simulation instruction corresponding to the current user and load the target role model corresponding to the role simulation instruction.

[0027] The role simulation command is used to trigger the current user to select multiple agents to be invoked, with each agent corresponding to a role model. For example, the role model corresponding to the test engineer agent is the test engineer model, and the role model corresponding to the project manager agent is the project manager model. More specifically, the intelligent agents can include: Module Implementation Planning Engineers, responsible for development environment configuration, task decomposition, development specification formulation, and risk prediction and response; Module Implementation Engineers, responsible for module coding implementation, API encapsulation, unit test writing, and commenting specifications; Module Test Tuning Engineers, responsible for unit / interface test design and execution, bug location and fixing, code / design refactoring, and performance optimization; Module Integration Acceptance Engineers, responsible for module-level integration, interface integration testing, functional demonstration, and review participation; Module Implementation Planning Mentors, responsible for guiding environment configuration and specification formulation, reviewing the rationality of task decomposition, guiding risk identification and analysis, and sharing practical experience and best practices; Module Implementation Guidance Mentors, responsible for reviewing key code and design, providing implementation method guidance, analyzing technical difficulties and solutions, and organizing / participating in code and design reviews; Test Tuning Guidance Mentors, responsible for guiding automated test design, guiding problem analysis and location, providing optimization and refactoring ideas, and reviewing test coverage and quality; and Module Integration Guidance Mentors, responsible for assisting in resolving cross-module interface issues and guiding integration test design. The roles and responsibilities of the engineers and engineers in the following roles are as follows: Design, organize / participate in final acceptance review, provide documentation standards and feedback; Development Engineers provide support for software coding, testing, and debugging; Development Engineers provide support for hardware implementation, testing, and debugging; Development Engineers provide support for data processing and model implementation; System Architects provide support for overall system architecture design; Module Designers provide detailed design support for individual modules; Product Managers define product requirements and features and evaluate implementation results; QA / Test Engineers design test strategies and test cases, execute tests, and report issues; DevOps Engineers set up development and testing environments and establish CI / CD processes; Integration Engineers coordinate interface connections and integration between different modules; Operations Engineers are responsible for system deployment and operation maintenance; Security Experts review security risks in code and design; Technical Documentation Engineers provide document templates and standards and review the quality of technical documentation; User Experience Designers provide user interface design specifications and evaluate user interaction experience; Data Engineers provide data model and data flow design.

[0028] After students log in to the simulated engineering practice education platform, they will generally first select an agent and then input their own needs or questions so that the selected agent can answer them. That is, when the current user's identity is obtained and the current user's identity is identified as a student, the system receives the current user's role simulation command, identifies the agent in the role simulation command, and loads the role model corresponding to each agent, thereby obtaining the target role model corresponding to the role simulation command.

[0029] Step S103: Receive the user requirements input by the current user, and based on the user requirements, generate the role response corresponding to each target role model, and generate the initial product requirement document based on the user requirements and role responses.

[0030] After loading the target role model corresponding to the current user, i.e., after the student completes the role model call, a text input box appears in the center of the operation interface of the simulated engineering practice education platform, allowing the student to input specific user requirements, such as "Design a smart fitness app for young people, requiring social functions." Upon receiving the student's user requirements, the electronic device distributes them to the loaded target role models. Each agent, based on its own role characteristics and built-in knowledge, uses natural language processing algorithms to generate corresponding role responses. For example, the product manager agent might provide "The overall architecture should include a user module, fitness course module, social interaction module, data analysis module, etc., prioritizing the usability of social functions," while the programmer agent might respond, "From a technical implementation perspective, the backend adopts a cloud server architecture, the frontend uses the ReactNative framework for cross-platform development, and the social functions utilize mature instant messaging APIs." The electronic device integrates these role responses and automatically generates an initial product requirements document according to a preset document template. This initial product requirements document covers project background, target users, functional requirements, and preliminary technical ideas, displayed in structured text format in the sidebar of the interface for easy viewing and modification by the student.

[0031] More specifically, one possible implementation of this application embodiment involves generating a role-based response for each target role model based on user needs, including: Identify key information in user needs, including at least one of product information, action information, and need information; Based on key information, determine the scope of responsibilities and interaction guidance degree corresponding to each target role model. The interaction guidance degree is a parameter that represents the ratio between the guiding behavior and the direct output behavior of the agent during the dialogue process. Based on the interaction guidance degree, responsibilities, and key information corresponding to each target role model, the corresponding intelligent agent domain knowledge base is triggered to generate a professional response, so as to obtain the role answer corresponding to each target role model.

[0032] Specifically, when an electronic device receives a user's input request, it first preprocesses the text content, such as removing irrelevant punctuation and special characters to make the text more organized. Then, using lexical and syntactic analysis tools from natural language processing, the text is broken down into words and phrases, and the grammatical relationships between them are analyzed. Next, using a pre-defined keyword database and semantic understanding model, the device identifies whether the text contains key information related to product information (such as product name, type, features, etc.), action information (such as development, design, optimization, etc.), and requirement information (such as functional requirements, performance requirements, time requirements, etc.). For example, if the user's request is "design a smartwatch," the electronic device will identify "smartwatch" as product information, "design" as action information, and no requirement information.

[0033] After identifying key information, the electronic device matches this information with the attributes and capabilities of each target role model. For each target role model (such as product manager, marketing manager, programmer, etc.), the electronic device pre-stores its corresponding scope of responsibilities. For example, a product manager's responsibilities include product planning, requirements analysis, and resource coordination; a programmer's responsibilities include coding, technical implementation, and debugging. By comparing the key information with the scope of responsibilities of each role model, the main responsibilities that each target role model should undertake in the current user demand scenario are determined.

[0034] While defining the scope of responsibilities, the electronic device determines the level of interaction guidance based on the type of the target role model and the complexity of the current key information. For complex requirements that require in-depth user thought and participation, such as the design of innovative product features, the product manager role model may have a higher level of interaction guidance, guiding users to think and discuss more rather than directly providing answers. Conversely, for relatively clear technical implementation problems, the programmer role model may directly output technical solutions with a lower level of interaction guidance. Specifically, when the completeness of user requirements is low (e.g., only describing the product name or core functions but lacking technical constraints), the level of interaction guidance increases, and the agent proactively guides users to supplement information through open-ended questions, option listings, or scenario assumptions. When the completeness of input information is high (e.g., a clear list of functions, user profiles, and performance metrics), the level of interaction guidance decreases, and the agent directly outputs technical solutions, architectural designs, or execution instructions. More specifically, the percentage of fields already filled in is calculated for the required fields of user requirements (product information, action information, and requirement information). For example: In the example above, the fields that have been filled in are product information and action information, and the fields that have not been filled in are requirement information. Therefore, the proportion of fields that have been filled in is 2 / 3, and the interaction guidance rate is 1 / 3.

[0035] Furthermore, based on the interaction guidance level, scope of responsibilities, and key information of each target role model, a corresponding domain knowledge base for the intelligent agent is located. The interaction guidance level determines how to retrieve information from the knowledge base and generate a response. Each target role model has its own exclusive domain knowledge base, storing the role's professional knowledge, experience, and case studies in the relevant domain.

[0036] Specifically, if the level of interactive guidance is high, electronic devices can extract guiding questions, ideas, or suggestions from the knowledge base to encourage further thinking and discussion among users. For example, a product manager role model might extract questions like "What are the key features of a smartwatch?" from the knowledge base to guide the user. If the level of interactive guidance is low, specific solutions, technical approaches, or suggestions matching the key information are directly retrieved from the knowledge base and presented in an appropriate language format. Furthermore, the information retrieved from the knowledge base is filtered and organized according to the role's responsibilities to ensure the output aligns with the role's duties in the current scenario. The generated professional responses are then presented to the user as the corresponding role answers for each target role model.

[0037] Step S104: Receive the user's decision input in real time and generate the corresponding decision effect.

[0038] After generating the initial product requirements document (PQD), it is displayed. Students, upon viewing this PQD, will input their requirements or questions again; these inputs can then be considered as user decisions. The electronic device receives these user decisions in real time and, using real-time data analysis and simulation technology, displays different project progress outcomes based on these decisions, thus providing feedback on the effectiveness of students' decisions. This allows students to understand the impact of their decisions throughout the entire project lifecycle. For example, if a student chooses an unreasonable technical solution during the architecture design phase, the electronic device will reflect the impact of this decision in the form of performance issues or quality defects during subsequent development and testing phases.

[0039] Specifically, one possible implementation of this application embodiment includes, in this step, receiving the user's input decision in real time and generating the decision effect corresponding to the user's decision, including: Receive the current user's decision; Based on user needs and user decisions, extract key decision parameters from each user's decision; Based on the pre-built engineering impact model, analyze the impact parameters of each user's decision on subsequent stages; Based on key decision parameters and influencing parameters, at least two virtual project progress paths are generated, corresponding to the engineering results under different decision combinations. Problem decision nodes are marked based on the engineering results, and decision effects are generated.

[0040] Upon receiving a user's decision, the input can be preprocessed, such as removing extra spaces and special characters to ensure proper formatting. Natural language processing (NLP) techniques are then used to analyze the text, extracting key information to determine crucial decision parameters, such as product information, action information, and requirement information. Simultaneously, real-time data analysis is performed, combining the initial product requirement document with other relevant data stored in the system (such as industry standards and historical project data). For example, the difference between the user's decision and the original requirements can be compared to assess the potential impact of the decision on project costs, timelines, and technical challenges.

[0041] Furthermore, pre-built engineering impact models can be invoked, and the extracted key decision parameters can be input into the corresponding engineering impact model. This engineering impact model can be a professional simulation model based on different engineering fields (such as software engineering, hardware engineering, and architectural engineering). For example, for software engineering projects, the engineering impact model might include sub-modules such as software architecture simulation, code writing simulation, and testing process simulation. Based on user decisions, the engineering impact model simulates the progress of the project at different stages and outputs the impact parameters of each user decision on subsequent stages, such as increased development difficulty, extended testing time, and increased probability of deployment risk, recorded in a quantitative or qualitative manner. During the simulation process, the engineering impact model considers the interaction and influence between various factors, such as the impact of technology selection on development efficiency and the impact of changes in development schedule on testing time. The simulation process can be presented to students visually, such as using animations and charts, allowing them to intuitively see how the project operates under different decisions.

[0042] Furthermore, based on the extracted key decision parameters and analyzed impact parameters, a path generation algorithm is used to generate at least two virtual project progress paths. For example, one path follows the user's current decision, while the other maintains the original plan. During path generation, the logical relationships and temporal sequence between stages are considered, simulating the actual progress of the project under different decision combinations. For each virtual project progress path, the simulation results can be used to determine if any problems or risks have arisen. If a decision node leads to project delays, cost overruns, or quality degradation, that decision node is marked as a problem decision node. The system displays the generated virtual project progress paths to the user in a visual manner, such as using flowcharts or Gantt charts. For each problem decision node, a detailed explanation is provided, including the type of problem, its cause, and its specific impact on the project. Simultaneously, a summary report of the decision's effectiveness is generated in text, charts, and other formats, clearly presenting the engineering results under different decision combinations and helping users intuitively understand the impact of their decisions on the project.

[0043] Step S105: Based on user decisions and the effects of those decisions, revise the initial product requirements document to obtain the revised product requirements document.

[0044] After viewing the decision-making effect display, if the current user (i.e., the student) believes that the current decision needs adjustment and optimization, they can directly modify the decision content in the original decision input box. The electronic device will then generate a new decision effect based on the received signal, compare the changes before and after, and update relevant chapters and clauses in the initial product requirements document using text processing algorithms based on the latest user decision and its effect. This includes adjusting the priority of functional modules, refining technical implementation details, and modifying project timelines. The revised complete product requirements document is then displayed again in the sidebar, replacing the original initial version, and document version change information is recorded for easy retrospective review. If the student has not entered any modified decision content, the initial product requirements document is directly supplemented based on the user's decision and its effect to obtain a revised product requirements document.

[0045] Step S106: Based on the effect of each decision and the revised product requirements document, conduct an independent evaluation of each user decision to obtain an independent evaluation result, and display the revised product requirements document and the independent evaluation result.

[0046] In addition to the vertical waterfall workflow, the simulated engineering practice education platform also includes a horizontal evaluation module. This module reviews each student's user decisions, generating analysis reports to guide students at each stage and provide improvement suggestions. The horizontal evaluation module covers multiple dimensions of expert consultation engineering capability certification: economic decision-making, environment, law, ethics, user perspective, etc. Through consultation, students can modify their engineering documents based on insightful and valuable expert advice. Specifically, after each student inputs a user decision and receives a revised product requirement document, the horizontal evaluation module scores and evaluates each decision from multiple dimensions, including but not limited to technical feasibility (e.g., whether the selected technology is mature and feasible), economic rationality (whether it exceeds the budget, resource utilization rate), time suitability (whether it meets project schedule requirements), and user need satisfaction (whether it meets initial user needs and subsequent adjustment needs). Each dimension is assigned a certain weight, and a weighted calculation yields an independent evaluation result for each user decision, presented as a score or grade (e.g., excellent, good, satisfactory, unsatisfactory).

[0047] Furthermore, the revised product requirements document is presented in its entirety, and a list of user decisions and their corresponding independent evaluation results is displayed on the right side of the document. This allows students to summarize their experiences and lessons learned, and provides a reference for the next round of simulation practice.

[0048] This application provides a simulated engineering practice education method that offers personalized experiences to students through identity recognition, loads multi-agent role models using role simulation instructions to help students understand engineering collaboration from different role perspectives, generates role responses and initial product requirement documents based on user needs to cultivate requirements analysis and management capabilities, receives user decisions in real time and generates decision effects to enhance feedback on decision practice, improves iterative optimization and critical thinking through document revision and independent decision evaluation, and finally visualizes the revised documents and evaluation results to promote knowledge integration and reflection, forming a complete practice loop and effectively improving students' comprehensive ability to solve complex engineering problems.

[0049] Another possible implementation of this application embodiment, after obtaining the current user's identity in step S101 above, the method further includes: When the user is a teacher, obtain the user's corresponding user instructions. The user instructions are used for one of the following: issuing training tasks, setting evaluation criteria, and manual score correction. When the user's instruction is for manual scoring correction, the training information corresponding to the current user is obtained and displayed. The training information includes each student's input content, the invoked agent, and the evaluation result. Receive the student rating input by the current user and add the student rating to the corresponding student's teacher rating.

[0050] Once the user's identity is confirmed as a teacher and the account and password are correct, the teacher enters the simulated engineering practice education platform. The platform interface displays a function option panel clearly listing three main function buttons: "Publish Training Task," "Set Evaluation Criteria," and "Manual Score Correction." Teachers can input user commands by clicking the corresponding buttons. When a teacher clicks a button, the electronic device displays a detailed operation window based on the received signal. For example, clicking the "Publish Training Task" button will bring up a window with input boxes for the task title, description, requirements, and deadline, where the teacher can fill in the relevant training task information. Clicking the "Set Evaluation Criteria" button will display an editable evaluation indicator system interface, allowing the teacher to add, modify, or delete evaluation indicators and set their weights. Clicking the "Manual Score Correction" button will take the teacher to the manual score correction operation page, awaiting further instructions.

[0051] Furthermore, upon receiving a user instruction from a teacher to select "human-assisted scoring correction," the system searches the backend database for training information on all students corresponding to that teacher. This training information includes requirement analysis documents, design plans, and decision-making records submitted at different stages. It also retrieves information on the agents invoked by students during training, such as which role models (product manager, programmer, etc.) were invoked, the number of invocations, and the timing. Additionally, it includes automatically generated evaluation results, including scores, grades, and related evaluation descriptions for each student's task completion. The electronic device organizes and categorizes this training information, clearly displaying it in tabular or list format on the user interface for easy viewing and comparison of each student's performance. Teachers can further view and manipulate the displayed information using scroll bars and filtering functions.

[0052] Furthermore, after displaying each student's training information, the electronic device controls the simulated engineering practice education platform to set up a rating input box next to each student's information entry. Teachers can enter their ratings for the student in this box. Ratings can be specific scores (e.g., percentages, 10 out of 10) or grades (e.g., excellent, good, average, pass, fail). After the teacher enters the rating, they click the "confirm" or "submit" button, and the electronic device receives the student's rating in real time. Then, the electronic device associates the rating with the corresponding student's identity information and stores it in the corresponding teacher rating field in the backend database. The electronic device also provides a rating record viewing function, allowing teachers to check their ratings at any time and modify or adjust them if necessary.

[0053] Furthermore, if the system receives a user command from the teacher to "publish training task," the electronic device will control the simulated engineering practice education platform to display a detailed task creation window. Teachers can select targeted questions from the system's built-in question bank, categorized according to different engineering competency dimensions, such as system analysis skills and project management skills. Teachers can also edit the question content themselves, clearly defining the training objective as improving one or more of the students' engineering skills. Simultaneously, teachers can set the task deadline, task description, and required deliverables. The system will synchronize this task information to the students' devices in real time, allowing students to see the new training task on their own interfaces.

[0054] If the teacher selects the user instruction "Set Evaluation Criteria," the electronic device will control the simulated engineering practice education platform to display an editable evaluation indicator system interface. Teachers can adjust the evaluation indicators and their weights according to different engineering capability dimensions. For example, for tasks emphasizing system design skills, teachers can increase the weight of system design-related indicators and control the simulated engineering practice education platform to store these evaluation criteria in the backend database for evaluation after students complete the task.

[0055] Another possible implementation of this application embodiment, after generating the role answer corresponding to each target role model in step S103 above, further includes: Obtain the specific functional tasks corresponding to each intelligent agent; Based on user needs and specific functional tasks, a recommendation agent is determined.

[0056] When the electronic device initializes on the simulated engineering practice education platform, it reads the exclusive functional task information corresponding to each agent from the built-in agent configuration database. The agent configuration database is pre-built and maintained based on the division of labor and responsibilities of different engineering roles in actual projects. For example, the exclusive functional tasks of the product manager agent include, but are not limited to, market research, competitor analysis, product planning, requirement definition, and prioritization; the exclusive functional tasks of the programmer agent cover code writing, software architecture design, algorithm optimization, debugging, and maintenance; and the exclusive functional tasks of the marketing manager agent include market positioning, target customer group analysis, brand promotion strategy formulation, and marketing channel expansion. When a user initiates a request, the agent configuration platform can accurately extract the exclusive functional task text description corresponding to each agent based on the currently active agent list and perform associated markings in the background for easy subsequent processing.

[0057] Furthermore, after obtaining the user's request text, natural language processing technology is used to analyze it and extract key requirement information, such as the type of product involved, the expected functional goals, the technical challenges faced, and the required marketing direction. This key requirement information is then matched with the specific functional tasks of each agent obtained in the previous step to calculate the degree of matching. Agents with a high degree of matching with the user's needs are identified as recommended agents and presented to the user on the operation interface in a way that highlights them (such as bright icons, bold fonts, etc.) or places them at the top of the recommendation list, so that the user can refer to the answers of these agents first to solve the problem.

[0058] In another possible implementation of this application embodiment, after obtaining the revised product requirements document in step S105 above, the method further includes: Receive the document generation instructions from the current user and determine the completeness of the revised product requirements document; Based on completeness, determine whether a final product requirements document can be generated; Once it is determined that the final product requirements document can be generated, the final product requirements document corresponding to the current user is generated.

[0059] Once students receive the revised product requirements document, a "Generate Final Document" button appears on the simulated engineering practice education platform. Clicking this button is considered issuing a document generation instruction, which the electronic device receives in real time. Upon receiving the instruction, the electronic device controls the simulated engineering practice education platform to initiate a completeness assessment process. Specifically, the platform checks the revised product requirements document against a pre-defined set of evaluation criteria. These criteria cover multiple aspects, such as whether the functional requirements are clear and comprehensive, including the core functions, auxiliary functions, and interaction logic between them; whether performance requirements are detailed, such as response time, throughput, and stability; whether user experience requirements are mentioned, such as interface friendliness and ease of operation; and whether a preliminary technical implementation plan exists, involving basic information such as the selected technical architecture and programming language. The platform scores each of these points and then calculates the document's completeness score.

[0060] Furthermore, if the completeness reaches or exceeds a pre-set threshold (e.g., a score of 80 or above corresponds to high completeness), the simulated engineering practice education platform determines that the final product requirement document can be generated. At this point, a prompt box will pop up on the simulated engineering practice education platform interface, informing the user that "the current document completeness meets the requirements and the final product requirement document can be generated." If the completeness does not meet the standard, the simulated engineering practice education platform will pop up a warning box, displaying "Document completeness is insufficient; the final product requirement document cannot be generated," and listing in detail the missing or weak parts, such as "Missing mobile adaptation performance requirements; please supplement," guiding the user back to the document revision stage to continue optimization.

[0061] Furthermore, once the simulated engineering practice education platform determines that the final product requirements document can be generated, the electronic device automatically starts the document generation process. Specifically, the simulated engineering practice education platform optimizes the format of the revised product requirements document, standardizing fonts, font sizes, paragraph spacing, etc., to make its appearance more standardized and professional. It also integrates the content according to a predetermined document template, ensuring that the order of each part is logical. For example, the opening section is a project overview, followed by a description of functional requirements, performance requirements, technical implementation solutions, project schedule, etc., thus obtaining the final product requirements document for the current user.

[0062] Another possible implementation of this application embodiment, after obtaining independent evaluation results for each user's decision in step S106 above, the method further includes: Based on user needs and user decisions, collect evidence of the current user's contribution to the product requirement document generation process; Call upon engineering training expert agents to generate agent scores and improvement suggestions based on multiple engineering capability dimensions; Receive teacher ratings, revise the agent ratings based on the teacher ratings to obtain a comprehensive rating, and output the comprehensive rating.

[0063] Among them, the evidence of contribution consists of relevant information and evidence recorded during the product requirement document generation process, which records the impact of users on the document content. This is used to measure the degree of user contribution in the process and is an important basis for evaluating the role of users in the project.

[0064] Specifically, as users input their needs and make a series of decisions, electronic devices record in real time the user's actions during the product requirement document generation process. These actions include text editing (such as adding / modifying functional or non-functional requirements), agent interaction (adopting suggestions or rejecting proposals), and decision-making moments (such as proactively proposing technical constraints during the requirements review phase). The system also calculates the percentage of newly added / modified requirement items in the complete requirement document (e.g., increasing from an initial 20 to 35 items, resulting in a 75% increase in coverage). Based on the newly added / modified requirement items and the coverage rate, evidence of the user's contribution to the product requirement document generation process is obtained.

[0065] Furthermore, after completing the collection of contribution evidence, the electronic device invokes the engineering training expert agent and, by integrating the evaluation results of multiple engineering training patent agents, generates an agent scoring report. This report uses a radar chart to visually present the student's score distribution across various engineering competency dimensions, accompanied by detailed textual explanations outlining strengths and areas for improvement, enabling students to understand their overall engineering literacy. The engineering training expert agents encompass multiple dimensions, including economic decision-making, environmental, legal, ethical, and user-centric perspectives. Each agent operates based on pre-built industry standards, professional knowledge systems, and evaluation algorithms. For example, the economic decision-making agent delves into the details of user decisions regarding resource allocation, cost budgeting, and benefit estimation. It combines market dynamics and industry cost averages with big data to quantitatively score the rationality of the user's economic decisions. Simultaneously, by comparing best practice cases, it analyzes the shortcomings of the user's existing strategies and provides practical improvement suggestions such as "Optimizing procurement channels can reduce raw material costs by approximately [X]%." The environmental intelligence agent focuses on the project's potential ecological impact, reviewing user solutions from the perspectives of energy consumption, waste emissions, and sustainable material selection. It provides scores based on green engineering standards and proposes energy-saving alternatives for high-energy-consuming aspects. The legal compliance intelligence agent compares the user design with current laws, regulations, and industry standards item by item, marking potential violations and providing rigorous legal interpretations to help students operate in compliance and ensure the project proceeds within the legal framework. The ethics intelligence agent examines whether the project involves ethical issues such as privacy violations, data misuse, and lack of social responsibility, guiding students to establish correct values. The user-perspective intelligence agent simulates end-user scenarios, experiencing product design from aspects such as interactive convenience, visual comfort, and ease of use, and proposing optimization paths to improve user satisfaction.

[0066] Furthermore, teachers can log in to the simulated engineering practice education platform to provide teacher evaluations. The platform captures the teacher's input scores in real time and integrates the teacher's score with the agent's score based on a preset weighting algorithm. For example, if the teacher's score has a weight of 40% and the agent's score has a weight of 60%, and the teacher's score is 85 points and the agent's overall score is 78 points, then the overall score = 85 × 40% + 78 × 60% = 80.8 points.

[0067] After obtaining the comprehensive score, the simulated engineering practice education platform can display the agent's score, teacher's score, and comprehensive score in a bar chart. It also includes improvement suggestions from the agent and teacher from different perspectives, which helps students to fully understand their own performance and accurately anchor their future efforts.

[0068] The above embodiments introduce a simulated engineering practice education method from the perspective of process flow. The following embodiments introduce a simulated engineering practice education device from the perspective of virtual modules or virtual units. For details, please refer to the following embodiments.

[0069] See Figure 2 The simulated engineering practice education device 20 may specifically include: an acquisition module 201, a loading module 202, a receiving module 203, a generation module 204, a modification module 205, and a display module 206, wherein: A simulated engineering practice education device 20, comprising: Module 201 is used to obtain the current user's identity, which is either a student or a teacher. The loading module 202 is used to receive the role simulation instruction corresponding to the current user when the user's identity is a student, and load the target role model corresponding to the role simulation instruction. The role simulation instruction is used to trigger the current user to select multiple intelligent agents to be called, and one intelligent agent corresponds to one role model. The receiving module 203 is used to receive the user requirements input by the current user, generate the role response corresponding to each target role model based on the user requirements, and generate the initial product requirement document based on the user requirements and role responses. The generation module 204 is used to receive the user's input decision in real time and generate the decision effect corresponding to the user's decision; Modify module 205 to revise the initial product requirements document based on user decisions and the effects of those decisions, resulting in a revised product requirements document. Display module 206 is used to independently evaluate each user's decision based on the effect of each decision and the revised product requirement document, obtain independent evaluation results, and display the revised product requirement document and the independent evaluation results.

[0070] In one possible implementation of this application embodiment, when the generation module 204 receives the user's input decision in real time and generates the decision effect corresponding to the user's decision, it is specifically used for: Receive the current user's decision; Based on user needs and user decisions, extract key decision parameters from each user's decision; Based on the pre-built engineering impact model, analyze the impact parameters of each user's decision on subsequent stages; Based on key decision parameters and influencing parameters, at least two virtual project progress paths are generated, corresponding to the engineering results under different decision combinations. Problem decision nodes are marked based on the engineering results, and decision effects are generated.

[0071] In one possible implementation of this application embodiment, the simulated engineering practice education device 20 further includes a first determining module, specifically used for: Receive the document generation instructions from the current user and determine the completeness of the revised product requirements document; Based on completeness, determine whether a final product requirements document can be generated; Once it is determined that the final product requirements document can be generated, the final product requirements document corresponding to the current user is generated.

[0072] In one possible implementation of this application embodiment, the simulated engineering practice education device 20 further includes a calling module, specifically used for: Based on user needs and user decisions, collect evidence of the current user's contribution to the product requirement document generation process; Call upon engineering training expert agents to generate agent scores and improvement suggestions based on multiple engineering capability dimensions; Receive teacher ratings, revise the agent ratings based on the teacher ratings to obtain a comprehensive rating, and output the comprehensive rating.

[0073] In one possible implementation of this application embodiment, when the receiving module 203 generates the role answer corresponding to each target role model based on user needs, it is specifically used for: Identify key information in user needs, including at least one of product information, action information, and need information; Based on key information, determine the scope of responsibilities and interaction guidance degree corresponding to each target role model. The interaction guidance degree is a parameter that represents the ratio between the guiding behavior and the direct output behavior of the agent during the dialogue process. Based on the interaction guidance degree, responsibilities, and key information corresponding to each target role model, the corresponding intelligent agent domain knowledge base is triggered to generate a professional response, so as to obtain the role answer corresponding to each target role model.

[0074] In one possible implementation of this application embodiment, the simulated engineering practice education device 20 further includes a second determining module, specifically used for: Obtain the specific functional tasks corresponding to each intelligent agent; Based on user needs and specific functional tasks, a recommendation agent is determined.

[0075] In one possible implementation of this application embodiment, the simulated engineering practice education device 20 further includes a scoring module, specifically used for: When the user is a teacher, obtain the user's corresponding user instructions. The user instructions are used for one of the following: issuing training tasks, setting evaluation criteria, and manual score correction. When the user's instruction is for manual scoring correction, the training information corresponding to the current user is obtained and displayed. The training information includes each student's input content, the invoked agent, and the evaluation result. Receive the student rating input by the current user and add the student rating to the corresponding student's teacher rating.

[0076] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] This application also describes an electronic device from the perspective of a physical device, such as... Figure 3 As shown, Figure 3 The illustrated electronic device 300 includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device 300 does not constitute a limitation on the embodiments of this application.

[0078] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0079] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0080] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0081] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0082] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers, and can also be servers, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0083] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0084] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0085] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A simulated engineering practice education method, characterized in that, include: Obtain the current user's identity, which is either a student or a teacher; When the user is a student, the system receives the role simulation instruction corresponding to the current user and loads the target role model corresponding to the role simulation instruction. The role simulation instruction is used to trigger the current user to select multiple intelligent agents to be invoked, with one intelligent agent corresponding to one role model. Receive the user requirements input by the current user, and based on the user requirements, generate the role response corresponding to each target role model, and generate an initial product requirement document based on the user requirements and the role response; The system receives user decisions input by the current user in real time and generates the corresponding decision effects. Based on the user's decision and the effect of the decision, the initial product requirement document is revised to obtain the revised product requirement document. Based on the effects of each decision and the revised product requirements document, each user's decision is independently evaluated to obtain an independent evaluation result, and the revised product requirements document and the independent evaluation result are displayed. The real-time reception of user decisions input by the current user and the generation of corresponding decision effects include: receiving the user decisions of the current user; extracting key decision parameters from each user decision based on the user's needs and the user decisions; analyzing the impact parameters of each user decision on subsequent stages according to a pre-built engineering impact model; generating at least two virtual project progress paths based on the key decision parameters and the impact parameters, corresponding to engineering results under different decision combinations, marking problem decision nodes based on the engineering results, and generating decision effects. The step of generating a role-based response for each target role model based on the user needs includes: identifying key information in the user needs, the key information including at least one of product information, action information, and demand information; determining the scope of responsibility and interaction guidance degree for each target role model based on the key information, the interaction guidance degree being a parameter representing the ratio between guiding behavior and direct output behavior of the agent during dialogue; and triggering the corresponding agent domain knowledge base to generate a professional response based on the interaction guidance degree, scope of responsibility, and key information for each target role model, so as to obtain a role-based response for each target role model.

2. The simulated engineering practice education method according to claim 1, characterized in that, After obtaining the revised product requirements document, the method further includes: receiving the document generation instruction from the current user and determining the completeness of the revised product requirements document; based on the completeness, determining whether a final product requirements document can be generated; and when it is determined that the final product requirements document can be generated, generating the final product requirements document corresponding to the current user.

3. The simulated engineering practice education method according to claim 1, characterized in that, After obtaining independent evaluation results for each user's decision, the method further includes: collecting evidence of the current user's contribution to the product requirement document generation process based on the user's needs and decisions; calling an engineering training expert agent to generate an agent score and improvement suggestions based on multiple engineering capability dimensions; receiving teacher scores and correcting the agent score based on the teacher scores to obtain a comprehensive score, and outputting the comprehensive score.

4. The simulated engineering practice education method according to claim 1, characterized in that, The method further includes: obtaining the exclusive functional tasks corresponding to each intelligent agent; and determining the recommended intelligent agent based on the user needs and the exclusive functional tasks.

5. The simulated engineering practice education method according to claim 1, characterized in that, After obtaining the current user's identity, the method further includes: when the user's identity is a teacher, obtaining the user instruction corresponding to the current user, the user instruction being used for one of issuing training tasks, setting evaluation criteria, and manual score correction; when the user instruction is manual score correction, obtaining the training information corresponding to the current user and displaying the training information, the training information including each student's input content, the invoked agent, and the evaluation result; receiving the student score input by the current user and adding the student score to the corresponding student's teacher score.

6. A simulated engineering practice education device, characterized in that, include: The acquisition module is used to acquire the user identity of the current user, which is either a student or a teacher. The loading module is used to receive the role simulation instruction corresponding to the current user when the user's identity is a student, and load the target role model corresponding to the role simulation instruction. The role simulation instruction is used to trigger the current user to select multiple intelligent agents to be invoked, and one intelligent agent corresponds to one role model. The receiving module is used to receive the user requirements input by the current user, and generate the role response corresponding to each target role model based on the user requirements, and generate an initial product requirement document based on the user requirements and the role responses. A generation module is used to receive user decisions input by the current user in real time and generate the corresponding decision effects. Specifically, the generation module is used to: receive the user decisions of the current user; extract key decision parameters from each user decision based on the user's needs and user decisions; analyze the impact parameters of each user decision on subsequent stages according to a pre-built engineering impact model; generate at least two virtual project progress paths based on the key decision parameters and the impact parameters, corresponding to engineering results under different decision combinations, and mark problem decision nodes based on the engineering results, and generate decision effects. The modification module is used to modify the initial product requirement document based on the user's decision and the effect of the decision, so as to obtain the modified product requirement document. The display module is used to independently evaluate each user decision based on the effect of each decision and the modified product requirement document, obtain an independent evaluation result, and display the revised product requirement document and the independent evaluation result; wherein, the receiving module is further specifically used to: identify key information in the user requirements, the key information including at least one of product information, action information, and requirement information; based on the key information, determine the scope of responsibility and interaction guidance degree corresponding to each target role model, the interaction guidance degree being a parameter of the ratio between guiding behavior and direct output behavior of the intelligent agent during the dialogue process; Based on the interaction guidance degree, scope of responsibility, and key information corresponding to each target role model, the corresponding intelligent agent domain knowledge base is triggered to generate a professional response to obtain the role answer corresponding to each target role model.

7. An electronic device, characterized in that, The electronic device includes: at least one processor; a memory; at least one application, wherein the at least one application is stored in the memory and configured to be executed by the at least one processor, said at least one application being configured to: perform the simulated engineering practice education method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in the computer, the computer is instructed to perform the simulated engineering practice education method according to any one of claims 1 to 5.

Images