Intelligent practical training system based on computer neural network

Abstract

The invention relates to the technical field of practical training teaching, and discloses an intelligent practical training system based on a computer neural network, and the system comprises a knowledge explanation unit which employs a neural network model to provide personalized theoretical knowledge explanation; the skill operation simulation unit is used for constructing a simulation model to realize virtualized practice of skill operation; and the practical training evaluation unit is used for accurately evaluating the practical training performance of the user by using a deep learning technology. Meanwhile, the system is provided with a data acquisition unit which is responsible for acquiring user learning behaviors, operation simulation and practical training result data and transmitting the data into each model to acquire an output result. And finally, an intelligent feedback unit provides personalized practical training feedback and suggestions for the user through a defined intelligent feedback algorithm according to a model output result. According to the system, theoretical knowledge explanation and skill operation simulation are integrated, intelligentization and individuation of practical teaching are realized, and the practical teaching effect is effectively improved.

Description

Technical Field

[0001] The invention relates to the technical field of practical training teaching, and in particular to an intelligent practical training system based on computer neural network. Background Art

[0002] In the traditional practical teaching environment, theoretical knowledge explanation and skill operation training are two core components. However, the existing teaching methods have obvious limitations in both aspects.

[0003] In terms of theoretical knowledge explanation, traditional teaching methods mainly rely on teachers’ oral teaching or written teaching materials. This method often lacks interactivity and personalization, and is difficult to adapt to the learning styles and needs of different students. Students passively accept knowledge and lack opportunities for active participation and thinking, resulting in poor learning results. At the same time, teachers have a heavy teaching burden and it is difficult to provide personalized explanations and guidance for each student’s specific situation.

[0004] In terms of skill operation training, traditional practical teaching requires practice with actual equipment or simulators. However, factors such as limited equipment, site restrictions, and safety risks severely restrict students' practice opportunities. Students often need to compete for equipment within a limited time and cannot get sufficient practice and feedback. In addition, some high-risk skill operations have safety hazards when practiced on actual equipment, which brings additional challenges to teaching.

[0005] With the rapid development of computer technology and artificial intelligence, neural network technology has been widely used in various fields, providing new ideas and methods for practical training. In the field of computer-assisted teaching, neural network models can provide learners with personalized learning experience and feedback by learning and analyzing large amounts of data. However, most of the current intelligent training systems focus on single theoretical knowledge explanation or skill operation simulation, lacking a comprehensive solution that organically combines the two. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent training system based on computer neural network to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: an intelligent training system based on computer neural network, the system comprising:

[0008] The knowledge explanation unit builds a knowledge explanation model to provide theoretical knowledge explanation related to practical training;

[0009] Skill operation simulation unit, which builds a skill operation simulation model to simulate the skill operation process in practical training;

[0010] The training evaluation unit builds a deep learning-based training evaluation model, which uses training data to evaluate the user's training performance;

[0011] The data acquisition unit collects the user's learning behavior data, operation simulation data and practical training results data, and respectively transfers the collected data to the knowledge explanation model, skill operation simulation model and practical training evaluation model to obtain the output results of the model;

[0012] The intelligent feedback unit defines the intelligent feedback algorithm, analyzes the model output results, and provides users with personalized training feedback and suggestions.

[0013] Preferably, the knowledge explanation model adopts a Transformer architecture, including:

[0014] Input layer: receiving training topics and related questions;

[0015] Encoding layer: Encodes the input information through the self-attention mechanism to generate contextual representation;

[0016] Decoding layer: Generates corresponding explanation content based on the context representation;

[0017] Output layer: Output theoretical knowledge related to the practical training topic.

[0018] Preferably, the training step of the knowledge explanation model includes:

[0019] Collect a large number of practical training topics and corresponding knowledge explanation texts as training data;

[0020] Take the training topic as input and the knowledge explanation text as label;

[0021] Train the Transformer model using the training data and optimize the model by minimizing the difference between the generated text and the real text;

[0022] Use the validation set to validate the model and evaluate the model's explanation accuracy and fluency;

[0023] When the model performance reaches the preset standard, the model parameters are saved to obtain the trained knowledge explanation model.

[0024] Preferably, the skill operation simulation model adopts a generative adversarial network (GAN) architecture, including:

[0025] Generator: receives operation instructions and generates simulated operation process data;

[0026] Discriminator: distinguishes the generated operation process data from the real operation process data;

[0027] Input layer: receives operation instructions input by users;

[0028] Output layer: outputs the simulated skill operation process.

[0029] Preferably, the training step of the skill operation simulation model includes:

[0030] Collect a large amount of real skill operation process data as training data;

[0031] The operation instructions are used as the input of the generator, and the actual operation process data is used as the positive sample of the discriminator;

[0032] Use the training data to train the GAN model and optimize the model by minimizing the discriminant error of the discriminator and the difference between the generator and the real data;

[0033] Use the validation set to validate the model and evaluate the authenticity and accuracy of the simulated operation process;

[0034] When the model performance reaches the preset standard, the model parameters are saved to obtain the trained skill operation simulation model.

[0035] Preferably, the training evaluation model adopts an architecture combining a convolutional neural network (CNN) and a long short-term memory (LSTM) network, including:

[0036] CNN part: Processes the user's learning behavior data and spatial features in the operation simulation data;

[0037] LSTM part: processes the time series features during user operations;

[0038] Fully connected layer: fuses the output features of CNN and LSTM and connects them to the output layer;

[0039] Output layer: Output the user's training evaluation results, including operation accuracy, completion time and step rationality.

[0040] Preferably, the training step of the training evaluation model includes:

[0041] Collect users’ learning behavior data, operation simulation data, and corresponding practical training evaluation results as training data;

[0042] Take learning behavior data and operation simulation data as input, and practical training evaluation results as labels;

[0043] Use the training data to train the CNN-LSTM model and optimize the model by minimizing the difference between the predicted evaluation results and the true evaluation results;

[0044] Use the validation set to validate the model and evaluate the model's accuracy and stability;

[0045] When the model performance reaches the preset standard, the model parameters are saved to obtain the trained training evaluation model.

[0046] Preferably, the implementation steps of the intelligent feedback algorithm defined in the intelligent feedback unit include:

[0047] Step 1: Define the output of the knowledge explanation model as K_Explain, the output of the skill operation simulation model as S_Simulate, and the output of the training evaluation model as E_Evaluate;

[0048] Step 2: Based on the results of E_Evaluate, identify the user’s weaknesses and areas for improvement in the training;

[0049] Step 3: Combined with K_Explain, provide users with supplementary explanations of theoretical knowledge targeting weaknesses;

[0050] Step 4: Combined with S_Simulate, provide users with skill operation simulation training targeting weaknesses;

[0051] Step 5: If there are multiple practical training projects or skill points, the above feedback process is carried out for each practical training project or skill point to obtain their own personalized feedback;

[0052] Step 6: Based on the personalized feedback of all training projects or skill points, integrate and generate the final training feedback report.

[0053] Preferably, the training feedback report also includes a user progress tracking mechanism, and the specific steps include:

[0054] S1: Record the user's evaluation results and feedback suggestions in each training;

[0055] S2: Compare the user's training performance at different time points and analyze the user's progress;

[0056] S3: According to the user's progress, adjust the content of knowledge explanation and the difficulty of skill operation simulation to adapt to the user's learning pace and ability level;

[0057] S4: In the training feedback report, show the user's progress trajectory and future learning suggestions.

[0058] Preferably, the data acquisition unit also includes a data enhancement module, which is responsible for performing data enhancement processing on the collected user learning behavior data, operation simulation data and training results data, including data amplification, data transformation and data synthesis.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The knowledge explanation unit in the system has built a knowledge explanation model, which can use neural network technology to provide theoretical knowledge explanation related to practical training. This method is not only rich in content and diverse in form, but also can be personalized according to students' learning styles and needs, enhancing the interactivity and fun of learning, thereby effectively improving students' learning effects. Through the intelligent feedback unit, the system can provide personalized practical training feedback and suggestions based on students' learning behavior data and performance, helping students better master knowledge and skills.

[0061] The skill operation simulation unit has built a skill operation simulation model that can simulate the skill operation process in actual training. This allows students to practice skill operation at any time and any place without relying on actual equipment or simulators, greatly breaking through the limitations of equipment quantity and venue. At the same time, since there is no safety risk in simulated operation, students can practice various high-risk skill operations with confidence, thereby obtaining more comprehensive practice opportunities and feedback.

[0062] The practical training evaluation unit has built a practical training evaluation model based on deep learning, which can use practical training data to objectively and accurately evaluate students' practical training performance. This helps teachers to have a more comprehensive understanding of students' learning situation and progress, so as to formulate more reasonable teaching plans and strategies.

[0063] Through the feedback of the practical training evaluation model, teachers can adjust teaching methods and content in a timely manner to improve teaching quality and effectiveness. The present invention can automatically perform knowledge explanation, skill operation simulation and practical training evaluation, which greatly reduces the teaching burden of teachers. Teachers can devote more energy to teaching research and innovation to improve teaching efficiency and quality. At the same time, the system can also provide teachers with personalized teaching suggestions and guidance based on students' learning data and performance, helping teachers to better meet students' learning needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a working principle diagram of the intelligent training system based on computer neural network described in the present invention;

[0065] Figure 2 A training flow chart for the skill operation simulation model;

[0066] Figure 3 A schematic diagram of the network structure design for the training evaluation model;

[0067] Figure 4 This is a diagram of the implementation steps of the intelligent feedback algorithm. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] See also Figure 1-4 The present invention provides a technical solution: an intelligent training system based on computer neural network, the system comprising:

[0070] Knowledge explanation unit:

[0071] Construct a knowledge explanation model based on neural network technology, which can understand and generate theoretical knowledge content related to practical training. During the model training process, a large amount of theoretical knowledge data, such as textbooks, teaching videos, lesson plans, etc., is used for learning and optimization to form a comprehensive and accurate knowledge base.

[0072] The knowledge explanation model dynamically generates personalized theoretical knowledge explanation content according to the user's learning progress and needs, supports multiple presentation forms such as text, audio, and video, and enhances the interactivity and fun of learning.

[0073] Skill operation simulation unit: construct a skill operation simulation model to simulate the skill operation process in practical training.

[0074] Practical training assessment unit:

[0075] Construct a deep learning-based training evaluation model, which uses training data (such as user operation records, achievement data, etc.) as input to evaluate the user's training performance.

[0076] The model can automatically identify and analyze the user's operation behavior through training and learning the standards and criteria of practical training evaluation, and give objective and accurate evaluation results. The evaluation results include multiple dimensions such as operation accuracy, efficiency, and safety, providing users with comprehensive practical training performance feedback.

[0077] Data acquisition unit:

[0078] The data acquisition unit is responsible for collecting the user's learning behavior data, operation simulation data, and practical training results data. By deploying the data acquisition module on the user side, various data generated by the user during the learning process, such as learning time, operation path, and result quality, are collected in real time. After preprocessing and formatting, the collected data is respectively passed to the knowledge explanation model, skill operation simulation model, and practical training evaluation model as model input data.

[0079] Intelligent Feedback Unit:

[0080] Define an intelligent feedback algorithm that analyzes and processes the model output results to generate personalized training feedback and suggestions. The algorithm dynamically adjusts the feedback content and form based on the user's learning progress, operation performance and evaluation results to meet the needs of different users. Feedback content includes knowledge point tips, operation skill suggestions, improvement directions, etc., which are presented to users in the form of text, audio or video to help users better master knowledge and skills.

[0081] The present invention will be further described below in conjunction with Examples 1 to 4:

[0082] Embodiment 1:

[0083] The knowledge explanation model adopts the Transformer architecture, including:

[0084] Input layer: Receives training topics and related questions as input. Training topics can be specific skill names, operation procedures or theoretical concepts, and related questions are specific questions raised by users regarding training topics or content that needs to be explained.

[0085] The input layer converts the training topics and related questions into numerical forms that the model can handle, such as word embedding vectors, for subsequent processing.

[0086] Encoding layer: The self-attention mechanism is used to encode the input information. The self-attention mechanism can capture the correlation between the elements in the input sequence and generate a representation containing contextual information. The encoding layer is composed of multiple encoder layers stacked together. Each encoder layer contains a self-attention sublayer and a feedforward neural network sublayer. The residual connection and layer normalization are used to enhance the expressiveness of the model.

[0087] Decoding layer: Based on the context representation generated by the encoding layer, the decoding layer generates the corresponding explanation content. The decoding layer also uses the self-attention mechanism, but on this basis, it also introduces the encoder-decoder attention mechanism to capture the association between the context information output by the encoding layer and the current generated content. The decoding layer is composed of multiple decoder layers stacked together, each of which contains a self-attention sublayer, an encoder-decoder attention sublayer, and a feedforward neural network sublayer.

[0088] Output layer: The output layer converts the representation generated by the decoding layer into natural language text, that is, the theoretical knowledge related to the training topic. The output layer usually uses the softmax function to generate the vocabulary probability distribution and generates the final text through strategies such as greedy search or beam search.

[0089] The training steps of the knowledge explanation model include:

[0090] Collect a large number of practical training topics and corresponding knowledge explanation texts as training data. These data can come from educational resources such as textbooks, teaching videos, and lesson plans, or can be obtained through expert annotation or crowdsourcing. Ensure that the training data covers a wide range of practical training topics and knowledge points to improve the generalization ability of the model.

[0091] Take the training topic as input and the knowledge explanation text as label. Perform preprocessing operations such as word segmentation and stop word removal on the input and label, and convert the text into word embedding vector form. Build a vocabulary table and assign a unique index to each word in the model for use in the training process.

[0092] The Transformer model is trained using the training data. During the training process, the model parameters are optimized by minimizing the difference between the generated text and the real text. The cross entropy loss function is used as the optimization target, and the model parameters are updated using optimization algorithms such as gradient descent.

[0093] Set appropriate hyperparameters such as learning rate and batch size to ensure the stability and convergence of model training.

[0094] Use the validation set to validate the model and evaluate the accuracy and fluency of the model's explanation. The validation set should contain practical training topics and knowledge points that are different from but related to the training set. Evaluate the performance of the model by calculating the similarity between the generated text and the real text (such as BLEU score) and manual evaluation.

[0095] When the model performance reaches the preset standard, the model parameters are saved to obtain a trained knowledge explanation model. The model parameters can be stored in a file or database for use in subsequent applications.

[0096] Assume that the training topic is "Operation process of CNC machine tools" and the related question is "How to set the processing parameters of CNC machine tools?" During the training process, the input layer receives "Operation process of CNC machine tools" and "How to set the processing parameters of CNC machine tools?" as input, and the encoding layer encodes the input information through the self-attention mechanism to generate a representation containing context information. The decoding layer generates the corresponding explanation content according to the context representation, such as "When setting the processing parameters of CNC machine tools, you first need to enter the parameter setting interface, then select the type of parameter to be set, such as feed speed, spindle speed, etc., and finally enter the specific value and save the setting." The output layer converts the explanation content into natural language text as the output of the model. Through training and optimization, the model can accurately and fluently answer questions about the operation process of CNC machine tools, providing users with personalized theoretical knowledge explanations.

[0097] Embodiment 2:

[0098] The skill operation simulation model in the present invention adopts a generative adversarial network (GAN) architecture, and the architecture of the model includes:

[0099] Generator: The generator receives operation instructions as input, which can be action sequences, parameter settings, or other control signals input by the user through interactive devices. The generator generates simulated operation process data based on the operation instructions. These data can be the movement trajectory, state changes, or sensor readings of the device in the virtual environment, which are used to simulate the real skill operation process.

[0100] Discriminator: The task of the discriminator is to distinguish the operation process data generated by the generator from the real operation process data. The real operation process data comes from the record of the actual skill operation process or the accurate simulation of the simulator. The discriminator compares the features of the generated data with the real data and outputs a discrimination result, which indicates the similarity between the generated data and the real data.

[0101] Input layer: The input layer is responsible for receiving operation instructions input by the user. These instructions can be obtained through various interactive methods, such as keyboard input, mouse clicks, voice commands, or gesture recognition. The input layer converts the operation instructions into numerical forms that can be processed by the model, such as vectors or matrices, so that the generator can perform subsequent processing.

[0102] Output layer: The output layer is responsible for outputting the simulated skill operation process. This can include motion animation of virtual devices, status display, or real-time update of sensor data. The output layer converts the operation process data generated by the generator into a form that the user can perceive so that the user can interact and practice.

[0103] The training steps of the skill operation simulation model include:

[0104] Collect a large amount of real skill operation process data as training data. This data can come from records of actual skill operation processes, such as videos, sensor data, or simulator logs. Ensure that the training data covers a wide range of skill operation scenarios and variations to improve the generalization ability of the model.

[0105] The operation instructions are used as the input of the generator, and the real operation process data is used as the positive sample of the discriminator. The input and samples are preprocessed as necessary, such as data cleaning, normalization or feature extraction. A data set is constructed to assign a unique identifier to each operation instruction and real operation process data in the model for use in the training process.

[0106] The GAN model is trained using the training data. During the training process, the model parameters are optimized by minimizing the discriminant error of the discriminator and the difference between the generator and the real data.

[0107] The goal of the discriminator is to accurately distinguish generated data from real data, while the goal of the generator is to generate simulated data that is as close to real data as possible to deceive the discriminator.

[0108] An alternating training strategy is adopted, that is, the discriminator is trained first, then the generator is trained, and this process is repeated until the model reaches a stable state.

[0109] Use the validation set to validate the model and evaluate the authenticity and accuracy of the simulated operation process. The validation set should contain skill operation scenarios and data that are different from but related to the training set. Evaluate the performance of the model by comparing the similarity between the simulated operation process and the real operation process, calculating error indicators, or performing manual evaluation.

[0110] When the model performance reaches the preset standard, the model parameters are saved to obtain a trained skill operation simulation model. The model parameters can be stored in a file or database for use in subsequent applications.

[0111] Assume that the skill operation is "the process of welding workpieces", and the operation instructions include the moving path of the welding gun, the setting of welding current, the control of welding speed, etc. During the training process, the input layer receives the operation instructions input by the user, such as "move the welding gun to the specified position on the workpiece, set the welding current to 100A, and weld at a speed of 5cm / s". The generator generates simulated welding process data according to the operation instructions, such as the movement trajectory of the welding gun in the virtual environment, the simulation effect of the welding arc, the temperature change of the workpiece, etc. The discriminator tries to distinguish these simulated data from the real welding process data (such as the actual welding process data recorded by high-precision sensors). Through training and optimization, the GAN model can generate more and more realistic welding process simulations, allowing users to practice and evaluate welding skills in a virtual environment. When the model performance reaches the preset standard, the model parameters are saved to obtain the trained skill operation simulation model for subsequent practical teaching.

[0112] Embodiment 3:

[0113] The training evaluation model uses a convolutional neural network (CNN) combined with a long short-term memory network (LSTM). The architecture of the model includes:

[0114] CNN part: The CNN part is responsible for processing the user's learning behavior data and spatial features in the operation simulation data. The learning behavior data can include the user's click, drag, zoom and other operation records during the training process, as well as the user's perspective movement, object selection and other behaviors in the virtual environment. The operation simulation data can be obtained from the skill operation simulation model, such as the data generated when the user performs welding, assembly and other operations in the virtual environment.

[0115] CNN extracts spatial features from these data, such as the location, shape, and texture of the operation, through structures such as convolutional layers and pooling layers.

[0116] LSTM part: The LSTM part is responsible for processing the time series features during the user operation process. It can capture the time sequence and dependency of user operations, such as which operation the user performed first, which operation the user performed later, and the time interval between these operations. Through its unique gating mechanism, LSTM can effectively process long time series data and avoid the problem of gradient disappearance or explosion.

[0117] Fully connected layer: The fully connected layer fuses the output features of CNN and LSTM. It concatenates or weights the spatial features extracted by CNN and the time series features extracted by LSTM to form a comprehensive feature representation. The fully connected layer maps the comprehensive features to the output layer through linear transformation and nonlinear activation function.

[0118] Output layer: The output layer outputs the user's training evaluation results. The evaluation results can include operation accuracy (such as the similarity between the user's operation and the standard operation), completion time (such as the time required for the user to complete the training task) and step rationality (such as the logic and sequence of the user's operation steps).

[0119] The output layer uses a softmax function or regression model to map the comprehensive features to specific evaluation indicators.

[0120] The training steps of the training evaluation model include:

[0121] Collect user learning behavior data, operation simulation data, and corresponding training evaluation results as training data. The data can come from the actual training process, obtained through sensors, log records, or simulators.

[0122] Take learning behavior data and operation simulation data as input, and actual training evaluation results as labels. Perform necessary preprocessing on the input data, such as data cleaning, normalization, feature extraction, etc. Build a data set and assign a unique identifier to each input data and label in the model for use in the training process.

[0123] Use the training data to train the CNN-LSTM model. During the training process, the model parameters are optimized by minimizing the difference between the predicted evaluation results and the actual evaluation results. The cross entropy loss function, mean square error loss function, or a custom loss function is used to measure the difference between the predicted results and the actual results.

[0124] Use optimization algorithms such as gradient descent and Adam to update model parameters until the model reaches convergence.

[0125] Use the validation set to validate the model and evaluate the model's evaluation accuracy and stability. The validation set should contain training scenarios and user behavior data that are different from but related to the training set. Evaluate the performance of the model by calculating the similarity, accuracy, recall, and other indicators between the predicted evaluation results and the actual evaluation results. At the same time, observe the performance of the model in different training scenarios and user behaviors to evaluate the stability and generalization ability of the model.

[0126] When the model performance reaches the preset standard, the model parameters are saved to obtain a trained training evaluation model. The model parameters can be stored in a file or database for use in subsequent applications. At the same time, the model training process, hyperparameter settings, and verification results are recorded to reproduce and further optimize the model.

[0127] Assuming the practical training task is "assembly of electronic components", the user's learning behavior data includes the user's operation records such as selecting electronic components, moving components to specified positions, and adjusting component directions in the virtual environment. The operation simulation data includes the user's gestures, strength, speed, and other simulation data during the assembly process. The practical training evaluation results include operation accuracy (such as whether the position of component assembly is accurate), completion time (such as the time required for the user to complete the assembly task), and step rationality (such as whether the user's assembly steps meet the process requirements).

[0128] During the training process, the learning behavior data and operation simulation data are used as input, and the actual training evaluation results are used as labels to train the CNN-LSTM model. The CNN part extracts the spatial features of user operations, such as the location and shape of components; the LSTM part captures the time series features of user operations, such as the order and time interval of operations. The fully connected layer fuses the output features of CNN and LSTM and connects them to the output layer. The output layer outputs the user's actual training evaluation results.

[0129] Through training and optimization, the CNN-LSTM model can accurately evaluate the user's performance in the electronic component assembly training, including the accuracy of the operation, the completion time, and the rationality of the steps. When the model performance reaches the preset standard, the model parameters are saved and the trained training evaluation model is obtained for subsequent training teaching and evaluation.

[0130] Embodiment 4:

[0131] The implementation steps of the intelligent feedback algorithm defined by the intelligent feedback unit include:

[0132] Step 1: Define model output

[0133] The output of the knowledge explanation model is defined as K_Explain, which contains the theoretical knowledge explanation content required by the user and is presented in the form of text, image or video.

[0134] The output of the skill operation simulation model is defined as S_Simulate, which provides users with simulation training for skill operations in a virtual environment, including operation steps, operation feedback, and simulation results.

[0135] The output of the training evaluation model is defined as E_Evaluate, which contains the evaluation results of the user during the training process, such as operation accuracy, completion time, step rationality and other indicators.

[0136] Step 2: Identify User Weaknesses

[0137] Based on the results of E_Evaluate, analyze the user's performance in the training and identify the skills or operation steps where the user has weaknesses or needs improvement. For example, the evaluation results show that the user's operation accuracy is low in the assembly step of a certain electronic component, or it takes too long to complete a task.

[0138] Step 3: Provide additional theoretical knowledge

[0139] Combined with K_Explain, it provides users with supplementary explanations of relevant theoretical knowledge for the weaknesses identified by users. For example, it provides detailed theoretical explanations of the structure, function, assembly requirements, etc. of the components for users’ weaknesses in electronic component assembly.

[0140] Step 4: Provide skill operation simulation training

[0141] Combined with S_Simulate, targeted skill operation simulation training is provided for users based on the weaknesses they have identified. For example, based on the user's weaknesses in the assembly steps, assembly simulation training in a virtual environment is provided, allowing users to practice repeatedly until they master the correct operation skills.

[0142] Step 5: Personalize the feedback process

[0143] If there are multiple practical training items or skill points, the above feedback process is performed for each practical training item or skill point separately.

[0144] For each practical training project or skill point, the weaknesses are identified based on the results of E_Evaluate, supplementary explanations of theoretical knowledge are provided in combination with K_Explain, and skill operation simulation training is provided in combination with S_Simulate. This ensures that users can get personalized feedback and training on each practical training project or skill point.

[0145] Step 6: Generate training feedback report

[0146] The final training feedback report is generated based on the personalized feedback of all training projects or skill points. The training feedback report includes the user's evaluation results on each training project or skill point, weakness identification, theoretical knowledge supplementary explanation content, skill operation simulation training status and improvement suggestions.

[0147] In the practical training feedback report, the steps to implement the user progress tracking mechanism include:

[0148] S1: In each training session, record the user's evaluation results and feedback suggestions. The evaluation results include indicators such as operation accuracy, completion time, and step rationality. The feedback suggestions include supplementary explanations of theoretical knowledge and suggestions for skill operation simulation training.

[0149] S2: Compare the user's training performance at different time points and analyze the user's progress in various training items or skill points. For example, compare the evaluation results of the user in the first training and after several training sessions, and observe the user's improvement in operation accuracy, completion time, etc.

[0150] S3: Adjust the content of knowledge explanation and the difficulty of skill operation simulation according to the user's progress.

[0151] If the user has mastered the relevant knowledge and demonstrated good operational skills at a certain skill point, the difficulty of theoretical knowledge explanation and skill operation simulation training for that skill point can be reduced.

[0152] On the contrary, if the user still has weaknesses or needs improvement in a certain skill point, the difficulty of theoretical knowledge explanation and skill operation simulation training for that skill point can be increased.

[0153] S4: In the training feedback report, the user's progress track is displayed, including the changes in the evaluation results of the user in each training project or skill point, the improvement of weaknesses, etc. At the same time, according to the user's progress and current learning level, the user is provided with future learning suggestions, such as which skill points the user should continue to strengthen and which new theoretical knowledge the user should learn.

[0154] Assume that a user is doing electronic assembly training, and the training project includes the assembly of resistors, capacitors, and integrated circuits. In the first training, the user's evaluation results show that the operation accuracy in the resistor assembly step is low and the completion time is long. Based on the evaluation results, the intelligent feedback unit identifies the user's weaknesses in the resistor assembly skill point.

[0155] Therefore, the intelligent feedback unit combines the knowledge explanation model to provide users with detailed theoretical knowledge supplementary explanations on the structure, function, assembly requirements, etc. of resistors. At the same time, combined with the skill operation simulation model, it provides users with virtual environment simulation training for resistor assembly, allowing users to practice repeatedly until they master the correct operation skills.

[0156] In the subsequent training, the intelligent feedback unit continued to evaluate the user's training performance and recorded the evaluation results and feedback suggestions. By comparing the user's training performance at different time points, the intelligent feedback unit found that the user's operation accuracy in the resistor assembly skill point has improved and the completion time has also been shortened.

[0157] Therefore, the intelligent feedback unit adjusted the content of knowledge explanation and the difficulty of skill operation simulation according to the user's progress. For the resistor assembly skill point, the difficulty of theoretical knowledge explanation and skill operation simulation training was reduced, and more emphasis was placed on the capacitor and integrated circuit assembly skill points.

[0158] Finally, the intelligent feedback unit integrates and generates the final training feedback report, which shows the user's progress trajectory in each training project or skill point and future learning suggestions. Users can understand their learning situation and progress trajectory based on the training feedback report, and carry out subsequent learning and training in a targeted manner.

[0159] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0160] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent training system based on computer neural network, characterized in that: The system comprises: The knowledge explanation unit builds a knowledge explanation model to provide theoretical knowledge explanation related to practical training; Skill operation simulation unit, which builds a skill operation simulation model to simulate the skill operation process in practical training; The training evaluation unit builds a deep learning-based training evaluation model, which uses training data to evaluate the user's training performance; The data acquisition unit collects the user's learning behavior data, operation simulation data and practical training results data, and respectively transfers the collected data to the knowledge explanation model, skill operation simulation model and practical training evaluation model to obtain the output results of the model; The intelligent feedback unit defines the intelligent feedback algorithm, analyzes the model output results, and provides users with personalized training feedback and suggestions.

2. According to claim 1, an intelligent training system based on computer neural network is characterized in that: The knowledge explanation model adopts the Transformer architecture, including: Input layer: receiving training topics and related questions; Encoding layer: Encodes the input information through the self-attention mechanism to generate contextual representation; Decoding layer: Generates corresponding explanation content based on the context representation; Output layer: Output theoretical knowledge related to the practical training topic.

3. The intelligent training system based on computer neural network according to claim 2 is characterized in that: The training steps of the knowledge explanation model include: Collect a large number of practical training topics and corresponding knowledge explanation texts as training data; Take the training topic as input and the knowledge explanation text as label; Train the Transformer model using the training data and optimize the model by minimizing the difference between the generated text and the real text; Use the validation set to validate the model and evaluate the model's explanation accuracy and fluency; When the model performance reaches the preset standard, the model parameters are saved to obtain the trained knowledge explanation model.

4. The intelligent training system based on computer neural network according to claim 1 is characterized in that: The skill operation simulation model adopts a generative adversarial network (GAN) architecture, including: Generator: receives operation instructions and generates simulated operation process data; Discriminator: distinguishes the generated operation process data from the real operation process data; Input layer: receives operation instructions input by users; Output layer: outputs the simulated skill operation process.

5. The intelligent training system based on computer neural network according to claim 4 is characterized in that: The training steps of the skill operation simulation model include: Collect a large amount of real skill operation process data as training data; The operation instructions are used as the input of the generator, and the actual operation process data is used as the positive sample of the discriminator; Use the training data to train the GAN model and optimize the model by minimizing the discriminant error of the discriminator and the difference between the generator and the real data; Use the validation set to validate the model and evaluate the authenticity and accuracy of the simulated operation process; When the model performance reaches the preset standard, the model parameters are saved to obtain the trained skill operation simulation model.

6. The intelligent training system based on computer neural network according to claim 1 is characterized in that: The training evaluation model adopts the architecture of combining convolutional neural network (CNN) and long short-term memory (LSTM) network, including: CNN part: Processes the user's learning behavior data and spatial features in the operation simulation data; LSTM part: processes the time series features during user operations; Fully connected layer: fuses the output features of CNN and LSTM and connects them to the output layer; Output layer: Output the user's training evaluation results, including operation accuracy, completion time and step rationality.

7. The intelligent training system based on computer neural network according to claim 6 is characterized in that: The training steps of the training evaluation model include: Collect users’ learning behavior data, operation simulation data, and corresponding practical training evaluation results as training data; Take learning behavior data and operation simulation data as input, and practical training evaluation results as labels; Use the training data to train the CNN-LSTM model and optimize the model by minimizing the difference between the predicted evaluation results and the true evaluation results; Use the validation set to validate the model and evaluate the model's accuracy and stability; When the model performance reaches the preset standard, the model parameters are saved to obtain the trained training evaluation model.

8. The intelligent training system based on computer neural network according to claim 1 is characterized in that: The implementation steps of the intelligent feedback algorithm defined in the intelligent feedback unit include: Step 1: Define the output of the knowledge explanation model as K_Explain, the output of the skill operation simulation model as S_Simulate, and the output of the training evaluation model as E_Evaluate; Step 2: Based on the results of E_Evaluate, identify the user’s weaknesses and areas for improvement in the training; Step 3: Combined with K_Explain, provide users with supplementary explanations of theoretical knowledge targeting weaknesses; Step 4: Combined with S_Simulate, provide users with skill operation simulation training targeting weaknesses; Step 5: If there are multiple practical training projects or skill points, the above feedback process is carried out for each practical training project or skill point to obtain their own personalized feedback; Step 6: Based on the personalized feedback of all training projects or skill points, integrate and generate the final training feedback report.

9. The intelligent training system based on computer neural network according to claim 8, characterized in that: The training feedback report also includes a user progress tracking mechanism, the specific steps include: S1: Record the user's evaluation results and feedback suggestions in each training; S2: Compare the user's training performance at different time points and analyze the user's progress; S3: According to the user's progress, adjust the content of knowledge explanation and the difficulty of skill operation simulation to adapt to the user's learning pace and ability level; S4: In the training feedback report, show the user's progress trajectory and future learning suggestions.

10. The intelligent training system based on computer neural network according to claim 1, characterized in that: The data acquisition unit also includes a data enhancement module, which is responsible for performing data enhancement processing on the collected user learning behavior data, operation simulation data and training achievement data, including data amplification, data transformation and data synthesis.

Images