Industrial autoclave intelligent training and examination system based on visual-touch collaborative perception

Abstract

The present application discloses an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception, which relates to the field of intelligent training technology, and solves the technical problem that the timeliness and effectiveness of the existing autoclave training and assessment system are low, resulting in low overall efficiency of the autoclave training system; including: a data acquisition module: used to obtain practical data, environmental data and face recognition data; a data processing module: constructing an autoclave digital twin model based on autoclave test machine data; using historical practical data to train an artificial intelligence model to obtain a prediction result generation model; and obtaining real-time practical data and environmental data, selecting an autoclave digital twin model or a prediction result generation model as a practical auxiliary model based on the practical data and environmental data; generating practical instrument data and practical feedback signals based on the practical auxiliary model, practical data and environmental data; and improving the overall efficiency of the training and examination system.

Description

Technical Field

[0001] The present application belongs to the field of intelligent training technology, and relates to intelligent training technology, specifically an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception. Background Art

[0002] Intelligent training examination is a new model that uses artificial intelligence technology to achieve intelligent management of the examination process. It improves the accuracy, efficiency and fairness of the examination through functions such as personalized test questions, automated score evaluation, real-time monitoring of test behavior, and data analysis and prediction. It has been widely used in school education, vocational training, medical training and other fields. With the continuous advancement of artificial intelligence technology, it shows a trend of becoming more intelligent, efficient and fair.

[0003] The prior art (the utility model patent with announcement number CN208589164U) discloses an autoclave fault simulation training and assessment system, which is characterized in that it includes a D / A conversion circuit, a signal analysis circuit connected to a terminal device, a prompt circuit connected to the safety accessories, auxiliary pipelines and the main body of the autoclave simulator, and an electric control circuit respectively connected to the safety accessories, auxiliary pipelines and the main body of the equipment; the D / A conversion circuit is respectively connected to the signal analysis circuit, the prompt circuit and the electric control circuit; the signal analysis circuit transmits the received analog instructions sent by the terminal device to the D / A conversion circuit; the D / A conversion circuit transmits the analog instructions to the electric control circuit and the prompt circuit; the electric control circuit is based on The autoclave is controlled to simulate corresponding faults according to the simulation instructions; the prompt circuit makes corresponding prompts according to the simulation instructions. The user practices repeatedly and masters the repair of various autoclave faults to ensure that when faced with an actual autoclave fault, reasonable emergency measures can be taken in time to avoid greater accidents; the above-mentioned autoclave fault simulation training and assessment system prompts trainees to perform corresponding operations during the assessment process by setting up a prompt circuit, so that trainees only need to perform corresponding operations according to the corresponding prompts during the training and assessment process; the mechanized training process leads to the low ability of trainees to handle problems independently; and ultimately leads to the low comprehensive training efficiency of the training and assessment system.

[0004] The prior art (invention patent announcement number CN115064020B) discloses a smart teaching method, system and storage medium based on digital twin technology, including: collecting entity data information of practical training teaching equipment, generating point cloud data of the practical training teaching equipment according to the entity data information, and generating a three-dimensional reconstruction model of the practical training teaching equipment through three-dimensional reconstruction of the point cloud data; establishing a digital twin model of the three-dimensional reconstruction model of the practical training teaching equipment, and constructing a virtual interactive scene according to the digital twin model; the target user conducts practical training and learning through an interactive terminal, obtains the target user's operational learning information during the training process, and evaluates the target user's practical training and learning status according to the operational learning information; formulates an adaptive practical training teaching plan according to the target user's practical training and learning status, and updates and corrects the adaptive practical training teaching plan through a periodic assessment system. The present invention improves the training efficiency by integrating the practical training teaching process with digital twins, and effectively solves the problems of low training efficiency and high risk factor. The intelligent teaching method of the above-mentioned digital twin model constructs a digital twin model of the training equipment to train each trainee according to their actual learning situation, thereby improving the training efficiency of the trainees. However, when the training equipment is more complex and there are more parameters to be considered in simulating the training equipment, the simulation time using the digital twin model will be longer. When more serious abnormalities occur, they actually need to be dealt with in a timely manner. However, due to the long simulation time of the digital twin model, the simulation results cannot be generated in time, resulting in a poor experience for the trainees when they are trained.

[0005] The existing intelligent training and assessment system has significant deficiencies in the training and assessment of autoclaves. Autoclaves require highly rigorous training due to the potential dangers during operation. However, the current training and assessment system is limited to providing preset questions, allowing trainees to operate the autoclave training and assessment equipment in a closed state according to the established steps; the defect of this approach is that the training equipment cannot display the actual state of the autoclave after each operation of the trainee in real time, resulting in the trainee only being able to mechanically memorize the operation steps and lacking the ability to solve problems independently; the system's effectiveness is low, resulting in low overall training efficiency. When trying to apply the digital twin model to the training and assessment of autoclaves, although it can theoretically simulate the real operating environment, due to the huge amount of data generated by the autoclave in actual operation, complex operations often cannot quickly produce results. This delay causes the practical teaching aids to be unable to provide timely feedback on the operation effect, the trainee experience is poor, and the training effect is greatly reduced. Especially in the face of emergency operation situations, the system cannot respond in time, resulting in the trainees being unable to respond in time when dealing with some more urgent operations; the system's timeliness is low, further weakening the comprehensive training efficiency of the training and assessment system; therefore, an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception is needed. Summary of the invention

[0006] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception, which is used to solve the technical problem that the timeliness and effectiveness of the existing autoclave training and assessment system are low, resulting in low overall efficiency of the autoclave training system.

[0007] To achieve the above-mentioned purpose, the first aspect of the present application provides an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception, including: a data acquisition module, an intelligent training and assessment module, a data processing module, an alarm module, a data library and a database;

[0008] The data acquisition module acquires real-time operation data, environmental data and face recognition data through the data acquisition device connected thereto;

[0009] The intelligent training and assessment module: obtains face recognition data, matches corresponding trainee information according to the face recognition data; obtains trainee information, generates training content and assessment content according to the trainee information; trains corresponding trainees according to the training content; assesses corresponding trainees according to the assessment data; obtains practical instrument data, and displays the assessment questions and practical instrument data at the corresponding positions of the autoclave test machine; and generates assessment results according to a number of practical data and practical instrument data;

[0010] The data processing module: obtains autoclave test machine data; constructs an autoclave digital twin model based on the autoclave test machine data; obtains a number of historical practical operation data through a database, and uses the historical practical operation data to train an artificial intelligence model to obtain a prediction result generation model; and,

[0011] Acquire real-time practical operation data and environmental data, and select the autoclave digital twin model or the prediction result generation model as the practical operation auxiliary model according to the practical operation data and environmental data; generate practical operation instrument data and practical operation feedback signals according to the practical operation auxiliary model, practical operation data and environmental data;

[0012] The alarm module generates an alarm signal according to the actual operation instrument data and the actual operation feedback signal, and issues an alarm according to the alarm signal.

[0013] The present application pre-builds a digital twin model of an autoclave and a prediction result generation model based on autoclave test machine data and a number of historical practical operation data; obtains trainee face recognition data during actual use; matches corresponding trainee information based on face recognition data; generates training content and assessment content based on trainee information; trains corresponding trainees based on the training content; assesses corresponding trainees based on assessment data; obtains practical operation data and environmental data in real time during training or assessment; selects a digital twin model of an autoclave or a prediction result generation model as a practical operation auxiliary model based on the practical operation data and environmental data; generates practical operation instrument data and practical operation feedback signals based on the practical operation auxiliary model, practical operation data and environmental data, and at the same time, generates an alarm signal based on the practical operation instrument data and practical operation feedback signals, and alarms; this embodiment performs the above operations; generates training content and assessment content matching the trainees based on their actual learning situation, thereby improving the trainees' experience; selects a suitable model as a practical operation auxiliary model based on the actual operation situation to enhance the timeliness and accuracy of the system response during training and assessment; enhances the timeliness and accuracy of the training and examination system; and thereby improves the overall efficiency of the training and examination system.

[0014] Preferably, the generating of training content and assessment content according to trainee information includes:

[0015] Extracting the theoretical scores and practical scores of each teaching stage from the student information; the theoretical scores and practical scores are the scores of the theoretical knowledge of autoclave and the actual operation of autoclave given by the teaching staff to each student according to the performance of the corresponding student at each stage; generating the theoretical status and practical status of the corresponding teaching stage according to the theoretical scores and practical scores;

[0016] When the theoretical status is the proficient status, the advanced theoretical courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage; otherwise, when the theoretical status is the intermediate status, the intermediate theoretical courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage; otherwise, the basic theoretical courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage;

[0017] When the practical operation status is a proficient status, the advanced practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; otherwise, when the practical operation status is a medium status, the medium practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; otherwise, the basic practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; the training content includes theoretical training content and practical training content;

[0018] Obtain the theoretical status and practical status of each teaching stage, set the proportion of questions with a proficient theoretical status to a high proportion; set the proportion of questions with a medium theoretical status to a medium proportion; set the proportion of questions with a basic theoretical status to a low proportion; randomly select a set number of questions from the theoretical test question sub-database of the database as theoretical assessment content according to the proportion of questions in each teaching stage;

[0019] The proportion of questions with proficient practical status is set to high proportion; the proportion of questions with medium practical status is set to medium proportion; the proportion of questions with basic practical status is set to low proportion; according to the proportion of questions in each teaching stage, a set number of questions are randomly selected from the practical test question sub-database of the database as practical assessment content; the assessment content includes theoretical assessment content and practical assessment content.

[0020] Preferably, generating the theoretical status and practical status of the corresponding teaching stage according to the theoretical score and the practical score includes:

[0021] Determine whether the theoretical score is greater than the high theoretical score threshold. If yes, set the theoretical status to the proficient status; if no, when the theoretical score is greater than the medium theoretical score threshold, set the theoretical status to the medium status; otherwise, set the theoretical status to the basic status;

[0022] Determine whether the practical score is greater than the advanced practical score threshold. If yes, set the practical status to the proficient status; if no, when the theoretical score is greater than the medium practical score threshold, set the practical status to the medium status; otherwise, set the practical status to the basic status.

[0023] Preferably, the digital twin model of the autoclave is constructed according to the autoclave test machine data, comprising:

[0024] Extract the autoclave test machine entity data and the autoclave use environment data from the autoclave test machine data; construct the autoclave three-dimensional model according to the autoclave test machine entity data, and construct the autoclave digital twin model according to the autoclave use environment data and the autoclave three-dimensional model;

[0025] Extract the operable parts in the autoclave test machine data, and establish a mapping relationship between the operable parts of the autoclave test machine and the corresponding operable parts of the autoclave digital twin model.

[0026] Preferably, the method of using historical practical operation data to train an artificial intelligence model to obtain a prediction result generation model includes:

[0027] Acquire several groups of historical practical operation data and the corresponding autoclave status labels and instrument data through the database; integrate several practical operation data in each group of practical operation data into a practical operation sequence according to their corresponding time, and obtain the time position of each practical operation data in the practical operation sequence; mark the sequence composed of the practical operation data before the corresponding time position of the corresponding practical operation sequence according to the time order as a partial practical operation sequence of the corresponding practical operation data;

[0028] Integrate a number of historical practical operation data, and the corresponding partial practical operation sequences and practical operation sequences, and the corresponding autoclave state labels and instrument data into a number of sets of training data and test data;

[0029] The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model whose input is a partial practical operation sequence and whose output is the corresponding autoclave state label and instrument data is obtained, the instrument data is used as predicted instrument data, the predicted instrument data and the autoclave state label are integrated into predicted result data for output, and finally a prediction result generation model is obtained; wherein the artificial intelligence model includes a BP neural network model and a RBF neural network model.

[0030] Preferably, the step of selecting the autoclave digital twin model or the prediction result generation model as the practical operation auxiliary model according to the practical operation data and the environmental data includes:

[0031] Acquire real-time practical operation data, and integrate various practical operation data from the beginning of this test or training to the present into a partial practical operation sequence in chronological order; generate a recommended use coefficient corresponding to the prediction result generation model according to the partial practical operation sequence;

[0032] Obtain real-time practical operation data and environmental data, and generate the recommended use coefficient 2 corresponding to the autoclave digital twin model based on the practical operation data and environmental data;

[0033] Determine whether the recommended use coefficient one is greater than the recommended use coefficient two; if yes, use the prediction result generation model as the practical auxiliary model; if no, use the autoclave digital twin model as the practical auxiliary model.

[0034] Preferably, the generating of the prediction result generation model according to the partial practical operation sequence corresponds to a recommended usage coefficient one, including:

[0035] Get the total data volume of some practical operation sequences, marked as ZS; get the reference throughput corresponding to some practical operation sequences, marked as TS; get the training sample volume XYL and the number of training sample types XYS used in the training process of the prediction result generation model, the test accuracy JZD in the test process, and the result accuracy SZD in the use process;

[0036] According to the training sample size XYL, the number of training sample types XYS, the test accuracy JZD and the result accuracy SZD, the recommended coefficient TJ1 of the corresponding prediction result generation model is generated by the following formula:

[0037]

[0038] Among them, α1 and α2 are proportional coefficients; β1, β2 and β3 are adjustment coefficients; γ1 and γ2 are weight coefficients; BYL is the basic training sample size; BYS is the number of basic training sample types.

[0039] Preferably, generating a recommended usage coefficient two corresponding to the autoclave digital twin model according to the practical operation data and the environmental data includes:

[0040] Get the value of each environmental item in the current environmental data, marked as HX i ; The influence coefficient SY of each practical project in the practical data j ; Obtain the simulation accuracy MZD of the autoclave digital twin model;

[0041] By formula The recommended coefficient 2 corresponding to the practical data and environmental data is calculated; among them, α3 and α4 are proportional coefficients; β4 and β5 are adjustment coefficients; i is the number of the environmental project, i=1, 2, ..., I; I is the total number of environmental projects; SHX i is the last corresponding value of the environmental project numbered i; j is the number of the practical project, j=1, 2,…, J; J is the total number of practical projects.

[0042] Preferably, the generating of the practical operation instrument data and the practical operation feedback signal according to the practical operation auxiliary model, the practical operation data and the environmental data includes:

[0043] Determine whether the practical auxiliary model is a prediction result generation model;

[0044] If yes, then obtain the practical operation data, integrate the practical operation data from the beginning of this test or training to the present into a partial practical operation sequence in chronological order; input the practical operation sequence into the prediction result generation model to obtain the prediction result data, and use the prediction instrument data in the prediction result data as the practical operation instrument data; when the autoclave status label in the prediction result data is abnormal, set the practical operation feedback signal to abnormal; otherwise, set the practical operation feedback signal to normal;

[0045] If not, obtain the actual operation data and environmental data, input the actual operation data and environmental data into the digital twin model of the autoclave, obtain the corresponding simulation result data, and use the simulated instrument data in the simulation result data as the actual operation instrument data; when the state of the autoclave in the simulation result data is abnormal, set the actual operation feedback signal to abnormal; otherwise, set the actual operation feedback signal to normal; the actual operation feedback signal includes abnormal or normal. It can be understood that the actual operation feedback signal is actually the state of the autoclave after the current actual operation.

[0046] Preferably, the step of generating an alarm signal according to the actual operation instrument data and the actual operation feedback signal comprises:

[0047] When the actual operation feedback signal is abnormal, the instrument value of each instrument in the actual operation instrument data is extracted, and each instrument value that exceeds the corresponding set safety range of the instrument is set as an abnormal instrument value, and an alarm signal of the abnormal instrument value is generated; otherwise, no alarm signal is generated.

[0048] Compared with the prior art, the beneficial effects of this application are:

[0049] 1. The present application pre-builds an autoclave digital twin model and a prediction result generation model based on the autoclave test machine data and some historical practical data; obtains practical data and environmental data in real time during training or assessment; selects an autoclave digital twin model or a prediction result generation model as a practical auxiliary model based on the practical data and environmental data; generates practical instrument data and practical feedback signals based on the practical auxiliary model, the practical data and the environmental data; selects a suitable model as a practical auxiliary model according to the actual operating conditions to enhance the timeliness and accuracy of the system response during training and assessment; enhances the timeliness and accuracy of the training and examination system; and thereby improves the overall efficiency of the training and examination system.

[0050] 2. This embodiment obtains the facial recognition data of students during actual use; matches the corresponding student information according to the facial recognition data; generates training content and assessment content according to the student information; trains the corresponding students according to the training content; assesses the corresponding students according to the assessment data; generates training content and assessment content that matches the students according to their actual learning situation, thereby improving the students' sense of experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 This is a schematic diagram of the principle of the autoclave smart training and examination system in this application;

[0053] Figure 2 Schematic diagram of the execution steps of the autoclave intelligent training and examination system in this application. DETAILED DESCRIPTION

[0054] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0055] See also Figure 1-Figure 2 , the first aspect of the present application provides an industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception, including: a data acquisition module, an intelligent training and assessment module, a data processing module, an alarm module and a database;

[0056] Data acquisition module: acquires real-time practical operation data, environmental data and face recognition data through the data acquisition equipment connected to it; the practical operation data refers to the operations performed by the trainees on the autoclave training and examination system, including the rotation angle of each valve, the pressing pressure of each button and the interval time of each operation, etc.; the environmental data refers to the data collected in the environment where the autoclave training and examination system is located, including the temperature, humidity and air pressure of the environment, as well as some data inside the autoclave, etc.; the face recognition data refers to the face image of the corresponding trainee collected when the trainee is undergoing training or assessment; the data acquisition equipment includes pressure acquisition equipment, temperature acquisition equipment and image acquisition equipment, etc.;

[0057] Intelligent training and assessment module: obtain face recognition data, match the corresponding trainee information according to the face recognition data, match the trainee in the image library according to the face image in the face data, query the trainee information of the corresponding trainee according to the matching result, and the trainee information includes the theoretical score and practical score of the corresponding trainee at each stage; obtain the trainee information, and generate training content and assessment content according to the trainee information; the training content is the training content generated according to the actual learning status of the corresponding trainee for training the trainee; the assessment content is the assessment content obtained by adjusting the set ratio according to the actual learning status of the trainee; train the corresponding trainees according to the training content; assess the corresponding trainees according to the assessment data; and obtain the practical instrument data, and display the assessment questions and practical instrument data at the corresponding position of the autoclave test machine; and, integrate the assessment result data according to a number of practical data and practical instrument data, and send the assessment result data to the relevant scoring department, and the relevant personnel of the scoring department score the trainee's assessment process according to the practical data and practical instrument data;

[0058] Data processing module: obtaining autoclave test machine data; the autoclave test machine data includes the test question data of the autoclave test machine and the use environment data of the autoclave test machine, and constructing the autoclave digital twin model according to the autoclave test machine data; obtaining a number of historical practical operation data through the database, the historical practical operation data is the data generated by the autoclave during the actual use process, including practical operation data, autoclave status labels and instrument data, etc.; using the historical practical operation data to train the artificial intelligence model to obtain a prediction result generation model; and,

[0059] During the practical assessment or practical training process; real-time practical data and environmental data are obtained, and the digital twin model of the autoclave or the prediction result generation model is selected as the practical auxiliary model according to the practical data and environmental data; the practical auxiliary model is an auxiliary computer model used to simulate the autoclave physical machine during the assessment or training process; practical instrument data and practical feedback signals are generated according to the practical auxiliary model, practical data and environmental data; the practical instrument data are the display data of each instrument after the corresponding actual operation, and the practical feedback signal is the state of the autoclave after the corresponding actual operation, including normal and abnormal;

[0060] Alarm module: Generates alarm signals based on actual operation instrument data and actual operation feedback signals, and issues alarms based on the alarm signals.

[0061] In this embodiment, a digital twin model of an autoclave and a prediction result generation model are pre-built according to the data of an autoclave test machine and a number of historical practical operation data; facial recognition data of trainees are obtained during actual use; corresponding trainee information is matched according to the facial recognition data; training content and assessment content are generated according to the trainee information; corresponding trainees are trained according to the training content; corresponding trainees are assessed according to the assessment data; during the training or assessment process, practical operation data and environmental data are obtained in real time; a digital twin model of an autoclave or a prediction result generation model is selected as a practical operation auxiliary model according to the practical operation data and environmental data; practical operation instrument data and practical operation feedback signals are generated according to the practical operation auxiliary model, the practical operation data and the environmental data, and at the same time, an alarm signal is generated according to the practical operation instrument data and the practical operation feedback signal, and an alarm is sounded; in this embodiment, through the above operations; training content and assessment content matching the trainees are generated according to the actual learning situation of the trainees, thereby improving the trainees' experience; by selecting a suitable model as a practical operation auxiliary model according to the actual operation situation, the timeliness and accuracy of the system response during the training and assessment process are enhanced; the timeliness and accuracy of the training and examination system are enhanced; and the comprehensive efficiency of the training and examination system is thereby improved.

[0062] When using the digital twin model to obtain the corresponding results, the operating data and environmental data of the corresponding equipment need to be input into the digital twin model. The digital twin model performs virtual operation. In the early stage of operation, too much data needs to be processed, resulting in a slow response time. The prediction result generation model trained by artificial intelligence has a shorter time series for prediction in the early stage, and the prediction result is generated faster. The later practical operation sequence is longer, resulting in a longer running time for the prediction result generation model. Therefore, this paper conducts a comprehensive evaluation of the above two models to determine which model to use as the practical auxiliary model for the autoclave training and examination machine, so as to enhance the timeliness and accuracy of the training and examination system.

[0063] Generating training content and assessment content according to trainee information, including: extracting theoretical scores and practical scores of each teaching stage in the trainee information; the theoretical scores and practical scores are the scores of each student's autoclave theoretical knowledge and autoclave practical operation by the teaching staff according to the corresponding trainee's performance at each stage; generating theoretical status and practical status of the corresponding teaching stage according to the theoretical scores and practical scores; the theoretical status is the learning status of the theoretical knowledge of the trainee at the corresponding teaching stage; the theoretical status is the practical operation status of the trainee at the corresponding teaching stage;

[0064] When the theoretical status is proficient, it means that the theoretical knowledge of the corresponding students in this teaching stage is very solid, and it is necessary to select training courseware with higher difficulty for training; then the advanced theoretical courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage; otherwise, when the theoretical status is medium, the medium theoretical courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage; otherwise, the basic theoretical courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage;

[0065] When the practical operation status is a proficient status, it means that the actual operation of the corresponding trainee in this teaching stage is very solid, and it is necessary to select a more difficult training courseware for training; then the advanced practical operation courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; otherwise, when the practical operation status is a medium status, the medium practical operation courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; otherwise, the basic practical operation courseware of the corresponding teaching stage is queried in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; the training content includes theoretical training content and practical training content;

[0066] The theoretical status and practical status of each teaching stage are obtained, and the proportion of questions with a proficient theoretical status is set to a high proportion; the proportion of questions with a medium theoretical status is set to a medium proportion; the proportion of questions with a basic theoretical status is set to a low proportion; according to the proportion of questions in each teaching stage, a set number of questions are randomly selected from the theoretical test question sub-database of the database as theoretical assessment content; the high proportion, the medium proportion and the low proportion can be proportion values ​​set in advance, the high proportion is greater than the medium proportion, and the medium proportion is greater than the low proportion; specific proportion values ​​can also be obtained by other methods, but it is necessary to satisfy that the high proportion is greater than the medium proportion, and the medium proportion is greater than the low proportion;

[0067] The proportion of questions with proficiency in practical operation is set to high proportion; the proportion of questions with medium practical operation is set to medium proportion; the proportion of questions with basic practical operation is set to low proportion; according to the proportion of questions in each teaching stage, a set number of questions are randomly selected from the practical test question sub-database of the database as practical assessment content; the assessment content includes theoretical assessment content and practical assessment content;

[0068] Specifically; there are simple questions, medium questions and difficult questions in the assessment content; initially, the proportion of questions corresponding to each teaching stage in simple questions is the same; the proportion of questions corresponding to each teaching stage in medium questions is the same; the proportion of questions corresponding to each teaching stage in difficult questions is the same; according to the students' actual learning situation, the proportion of questions in different teaching stages is adjusted, and corresponding question types are randomly selected from simple questions, medium questions and difficult questions according to the proportion of questions; for example, a student's practical operation status in teaching stage one is proficient, the practical operation status in teaching stage two is medium, and the practical operation status in teaching stage three is medium; then the proportion of questions corresponding to each teaching stage in the corresponding simple questions can be 4:3:3; the proportion of questions corresponding to each teaching stage in medium questions can be 4:3:3; the proportion of questions corresponding to each teaching stage in difficult questions can be 4:3:3; during the exam, slightly increasing the proportion of questions in the projects that students are good at can facilitate the adjustment of students' mentality and enable them to perform normally.

[0069] Generate the theoretical status and practical status of the corresponding teaching stage according to the theoretical score and the practical score, including: judging whether the theoretical score is greater than the high-level theoretical score threshold, if yes, set the theoretical status to the proficient status; if no, when the theoretical score is greater than the medium-level theoretical score threshold, set the theoretical status to the medium status; otherwise, set the theoretical status to the basic status; the high-level theoretical score threshold and the medium-level theoretical score threshold are both set according to expert experience; if the theoretical status is the proficient status, it means that the corresponding trainee has a solid theoretical learning in the corresponding teaching stage; if the theoretical status is the medium status, it means that the corresponding trainee has a relatively solid theoretical learning in the corresponding teaching stage; if the theoretical status is the basic status, it means that the corresponding trainee is at the basic stage of theoretical learning in the corresponding teaching stage;

[0070] Determine whether the practical score is greater than the advanced practical score threshold. If yes, set the practical status to the proficient status; if not, when the theoretical score is greater than the medium practical score threshold, set the practical status to the medium status; otherwise, set the practical status to the basic status; the advanced practical score threshold and the medium practical score threshold are set according to expert inspection; if the practical status is the proficient status, it means that the corresponding student has a solid practical operation learning in the corresponding teaching stage; if the practical status is the medium status, it means that the corresponding student has a relatively solid practical operation learning in the corresponding teaching stage; if the practical status is the basic status, it means that the corresponding student is in the basic stage of practical operation learning in the corresponding teaching stage.

[0071] Constructing a digital twin model of the autoclave according to the autoclave test machine data, including: extracting the autoclave test machine entity data and the autoclave use environment data in the autoclave test machine data; constructing a three-dimensional model of the autoclave according to the autoclave test machine entity data, and constructing a digital twin model of the autoclave according to the autoclave use environment data and the autoclave three-dimensional model;

[0072] Extract the operable parts from the autoclave test machine data, and establish a mapping relationship between the operable parts of the autoclave test machine and the corresponding operable parts of the autoclave digital twin model; facilitate the synchronization of the examinee's operations on the test machine to the autoclave digital twin model, and facilitate the synchronization of the analog instrument data generated by the digital twin model to the corresponding instruments of the autoclave test machine; the operable parts include valves, switches, buttons, instruments, etc.

[0073] The prediction result generation model is obtained by training an artificial intelligence model using historical practical operation data, including: obtaining several groups of historical practical operation data and corresponding autoclave state labels and instrument data through a database; the autoclave state label is the state of the autoclave after the corresponding practical operation data is performed in the corresponding group of practical operation data, including a normal state and an abnormal state; the instrument data is the display data of each instrument after the corresponding practical operation data is performed in the corresponding group of practical operation data; several practical operation data in each group of practical operation data are integrated into a practical operation sequence according to their corresponding time, and the time position of each practical operation data in the practical operation sequence is obtained; a sequence composed of practical operation data before the corresponding time position of each practical operation data in the corresponding practical operation sequence is marked in time order as a partial practical operation sequence of the corresponding practical operation data;

[0074] Integrate a number of historical practical operation data, and the corresponding partial practical operation sequences and practical operation sequences, and the corresponding autoclave state labels and instrument data into a number of sets of training data and test data;

[0075] The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; specifically, part of the practical operation sequence in the test data is input into the trained artificial intelligence model; the corresponding autoclave state label and instrument data output are obtained, and it is judged whether the output autoclave state label is consistent with the autoclave state label recorded in the test data; and whether the output instrument data output is within the range of the corresponding instrument data recorded in the test data, if yes, it means that the group of test data has passed the test, and the next group of test data is tested; if not, the relevant parameters in the model are adjusted, and the test data is used again for testing until a set proportion of the test data passes the test; an artificial intelligence model with a part of the practical operation sequence as input and a corresponding autoclave state label and instrument data as output is obtained, the instrument data is used as the predicted instrument data, the predicted instrument data and the autoclave state label are integrated into the predicted result data for output, and finally a prediction result generation model is obtained; wherein the artificial intelligence model includes a BP neural network model, a RBF neural network model, a recurrent neural network model, and an LSTM model.

[0076] According to the actual operation data and environmental data, the autoclave digital twin model or the prediction result generation model is selected as the practical operation auxiliary model, including:

[0077] Acquire real-time practical data, and integrate the practical data from the beginning of this test or training to the present into a partial practical sequence in chronological order; generate a recommended coefficient one corresponding to the prediction result generation model according to the partial practical sequence; the recommended coefficient one is the recommendation degree obtained by comprehensive evaluation of the prediction result generation model corresponding to the current practical data. The larger the value of the recommended coefficient one is, the better the performance of the prediction result generation model is, and the probability of using the prediction result generation model as a practical auxiliary model is greater;

[0078] Acquire real-time practical operation data and environmental data, and generate the recommended use coefficient 2 corresponding to the autoclave digital twin model based on the practical operation data and environmental data; the recommended use coefficient 2 is the recommendation degree obtained based on the comprehensive evaluation of the autoclave digital twin model corresponding to the current practical operation data. The larger the value of the recommended use coefficient 2 is, the better the performance of the autoclave digital twin model is, and the probability of using the autoclave digital twin model as a practical auxiliary model is higher;

[0079] Determine whether the recommended use coefficient one is greater than the recommended use coefficient two; if yes, use the prediction result generation model as the practical auxiliary model; if no, use the autoclave digital twin model as the practical auxiliary model.

[0080] When students first start to use the autoclave test machine, the corresponding practical data is relatively small, and the amount of data in the generated practical sequence is also small. At this time, the comprehensive efficiency of using the prediction result generation model to predict the autoclave practical results is higher; after a certain amount of practical data has been accumulated, the amount of data in the practical sequence will increase, and at this time, the efficiency of using the digital twin model to simulate the autoclave practical results is higher; it is understandable that basically only when the autoclave test machine is first used, it is necessary to select one of the autoclave digital twin model and the prediction result generation model as the practical auxiliary model every time the student performs an abnormal practice; after many uses, it can basically be determined that when practicing within the previously set number of times, the prediction result generation model needs to be used as the practical auxiliary model; after exceeding this number, the digital twin model needs to be replaced as the practical auxiliary model.

[0081] The recommended coefficient for the prediction result generation model based on some practical sequences is one, including:

[0082] Get the total data volume of some practical sequences, marked as ZS; get the reference throughput corresponding to some practical sequences, which is the average throughput of some practical sequences with similar data volume when the test data and the previous prediction result generation model are run; it can be understood that the closer the running time of the selected test data and the previous prediction result generation model is to the current time, the higher the accuracy of the reference throughput; marked as TS; get the training sample volume XYL and the number of training sample types XYS used in the training process of the prediction result generation model, the test accuracy JZD in the test process, the test accuracy is the accuracy of the test data of the model during the test process; and the result accuracy SZD during use; the result accuracy is the correctness of the result of each run after the model is put into use;

[0083] According to the training sample size XYL, the number of training sample types XYS, the test accuracy JZD and the result accuracy SZD, the recommended coefficient TJ1 of the corresponding prediction result generation model is generated by the following formula:

[0084]

[0085] Among them, α1 and α2 are proportional coefficients, which are used to adjust the proportional relationship of different items. The specific values ​​are set according to experience; β1, β2 and β3 are adjustment coefficients, which are similar to the functions of proportional coefficients and are used to assist the proportional coefficients in adjusting the proportional relationship of different parameters. The specific values ​​are set according to experience; γ1 and γ2 are weight coefficients, which are used to adjust the weights of the two items. BYL is the basic training sample size; BYS is the number of basic training sample types.

[0086] This embodiment calculates the recommended usage coefficient one of the prediction result generation model through the above formula. When the ratio of the data volume corresponding to the actual operation sequence of this time to the corresponding throughput is larger, it means that the expected processing time is longer, and the corresponding recommended usage coefficient one is set smaller; when the prediction result generation model uses more training samples and more training sample types in the training process, the accuracy in the verification process and the accuracy in the subsequent use process is higher, indicating that the accuracy of the subsequent prediction results is higher; the higher the corresponding recommended usage coefficient one is set; the prediction result generation model is comprehensively considered through the above method to obtain the corresponding recommended usage coefficient one.

[0087] Generate the recommended usage factor 2 corresponding to the autoclave digital twin model based on practical data and environmental data, including:

[0088] Get the value of each environmental item in the current environmental data, marked as HX i ; Environmental items include temperature, air pressure and humidity, etc.; The influence coefficient SY of each practical item in the practical data j; The influence coefficient is the coefficient generated by the influence range of the corresponding practical project. The larger the influence range, the larger the corresponding influence coefficient; the specific value is pre-set according to expert experience; if a valve is only used to control one pipeline, its opening and closing is only related to this pipeline and can only affect one pipeline; its corresponding influence coefficient is small; and a valve is used to control multiple pipelines or other components, and its opening and closing will affect multiple pipelines or other components; its corresponding influence coefficient is set larger; obtain the simulation accuracy MZD of the digital twin model of the autoclave; the simulation accuracy is the accuracy of the simulation result obtained by using the existing historical practical data to test the digital twin model of the autoclave after it is generated;

[0089] By formula The recommended coefficient TJ2 corresponding to the practical data and environmental data is calculated; among them, α3 and α4 are proportional coefficients; they are used to adjust the proportional relationship of different items, and the specific values ​​are set according to experience; β4 and β5 are adjustment coefficients, which are similar to the role of the proportional coefficient and are used to assist the proportional coefficient in adjusting the proportional relationship of different parameters, and the specific values ​​are set according to experience; it can be understood that the recommended coefficient 1 and the recommended coefficient 2 are ultimately limited to the same range through the adjustment of each proportional coefficient and adjustment coefficient; i is the number of the environmental project, i=1, 2,…, I; I is the total number of environmental projects; SHX i is the value corresponding to the last environmental item numbered i; j is the number of the practical item, j=1, 2, …, J; J is the total number of practical items; it can be understood that, before the autoclave digital twin model starts to simulate, the values ​​of the various environmental items simulated by it can be the values ​​of the various environmental items at the end of the last operation of the autoclave digital twin model; it can also be the initial state, that is, SHX i =0; In this embodiment, the initial state is used as the value of each environmental item before the autoclave digital twin model simulation operation, that is, SHX i =0.

[0090] In this embodiment, the recommended usage coefficient 2 of the digital twin model of the autoclave is calculated through the above formula; when the environment corresponding to the current environment and the previous practical operation data changes greatly, the digital twin model of the autoclave needs to update more simulated environmental data, and the corresponding data processing volume is larger, and the corresponding recommended usage coefficient 2 is set smaller; when the impact coefficient of the corresponding practical project is larger, it means that the scope of impact after the actual operation is larger, the complexity of the digital twin model of the autoclave needs to be simulated is greater, and the processing time is longer, and the corresponding recommended usage coefficient 2 is set smaller; when the simulation accuracy of the digital twin model of the autoclave is greater, the corresponding recommended usage coefficient 2 is set larger.

[0091] Generate practical instrument data and practical feedback signals based on the practical auxiliary model, practical data and environmental data, including:

[0092] Determine whether the practical auxiliary model is a prediction result generation model;

[0093] If yes, then obtain the practical operation data, integrate the practical operation data from the beginning of this test or training to the present into a partial practical operation sequence in chronological order; input the practical operation sequence into the prediction result generation model to obtain the prediction result data, and use the prediction instrument data in the prediction result data as the practical operation instrument data; when the autoclave status label in the prediction result data is abnormal, set the practical operation feedback signal to abnormal; otherwise, set the practical operation feedback signal to normal;

[0094] If not, then obtain the actual operation data and environmental data, input the actual operation data and environmental data into the digital twin model of the autoclave, obtain the corresponding simulation result data, and use the analog instrument data in the simulation result data as the actual operation instrument data; when the state of the autoclave in the simulation result data is abnormal, the actual operation feedback signal is set to abnormal; otherwise, the actual operation feedback signal is set to normal; the actual operation feedback signal includes abnormal or normal; it can be understood that the actual operation feedback signal is actually the state of the autoclave after the current actual operation;

[0095] Generate an alarm signal according to the actual operation instrument data and the actual operation feedback signal, including: when the actual operation feedback signal is abnormal, extract the instrument value of each instrument in the actual operation instrument data, set each instrument value that exceeds the corresponding set safety range of the instrument as an abnormal instrument value, and generate an alarm signal for the abnormal instrument value; otherwise, do not generate an alarm signal; so that the present embodiment can generate an alarm signal in time.

[0096] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0097] How this application works:

[0098] The present application pre-builds a digital twin model of an autoclave and a prediction result generation model based on autoclave test machine data and a number of historical practical operation data; obtains trainee face recognition data during actual use; matches corresponding trainee information based on face recognition data; generates training content and assessment content based on trainee information; trains corresponding trainees based on the training content; assesses corresponding trainees based on assessment data; obtains practical operation data and environmental data in real time during training or assessment; selects a digital twin model of an autoclave or a prediction result generation model as a practical operation auxiliary model based on the practical operation data and environmental data; generates practical operation instrument data and practical operation feedback signals based on the practical operation auxiliary model, practical operation data and environmental data, and at the same time, generates an alarm signal based on the practical operation instrument data and practical operation feedback signals, and alarms; this embodiment performs the above operations; generates training content and assessment content matching the trainees based on their actual learning situation, thereby improving the trainees' experience; selects a suitable model as a practical operation auxiliary model based on the actual operation situation to enhance the timeliness and accuracy of the system response during training and assessment; enhances the timeliness and accuracy of the training and examination system; and thereby improves the overall efficiency of the training and examination system.

[0099] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. Industrial-grade autoclave intelligent training and examination system based on visual and touch collaborative perception, including: Data collection module, intelligent training and assessment module, data processing module, alarm module, data base and database; characterized in that, The data acquisition module is used to obtain real-time practical operation data, environmental data and face recognition data; The intelligent training and assessment module: matches the corresponding trainee information according to the face recognition data; generates training content and assessment content according to the trainee information; trains the corresponding trainees according to the training content; and assesses the corresponding trainees according to the assessment data; The data processing module: obtains autoclave test machine data; constructs an autoclave digital twin model based on the autoclave test machine data; obtains a number of historical practical operation data through a database, and uses the historical practical operation data to train an artificial intelligence model to obtain a prediction result generation model; and, Acquire real-time practical operation data and environmental data, and select the autoclave digital twin model or the prediction result generation model as the practical operation auxiliary model according to the practical operation data and environmental data; including: Obtain real-time practical data, and integrate the practical data from the beginning of this exam or training to the present into a partial practical sequence in chronological order; generate a recommended usage coefficient corresponding to the prediction result generation model based on the partial practical sequence; including: Get the total data volume of some practical operation sequences, marked as ZS; get the reference throughput corresponding to some practical operation sequences, marked as TS; get the training sample volume XYL and the number of training sample types XYS used in the training process of the prediction result generation model, the test accuracy JZD in the test process, and the result accuracy SZD in the use process; According to the training sample size XYL, the number of training sample types XYS, the test accuracy JZD and the result accuracy SZD, the recommended coefficient TJ1 of the corresponding prediction result generation model is generated by the following formula: ； Among them, α1 and α2 are proportional coefficients; β1, β2 and β3 are adjustment coefficients; γ1 and γ2 are weight coefficients; BYL is the number of basic training samples; BYS is the number of basic training sample types; Obtain real-time practical operation data and environmental data, and generate the recommended use coefficient 2 corresponding to the autoclave digital twin model based on the practical operation data and environmental data; including: Get the value of each environmental item in the current environmental data, marked as HX i ; The influence coefficient SY of each practical project in the practical data j ; Obtain the simulation accuracy MZD of the autoclave digital twin model; By formula The recommended coefficient TJ2 corresponding to the practical data and environmental data is calculated; among them, α3 and α4 are proportional coefficients; β4 and β5 are adjustment coefficients; i is the number of the environmental project, i=1, 2, ..., I; I is the total number of environmental projects; SHX i is the last corresponding value of the environmental item with change i; j is the number of the practical item, j=1, 2,…, J; J is the total number of practical items; Determine whether the recommended use coefficient 1 is greater than the recommended use coefficient 2; if yes, use the prediction result generation model as the practical operation auxiliary model; if no, use the autoclave digital twin model as the practical operation auxiliary model; Generate practical instrument data and practical feedback signals according to the practical auxiliary model, practical data and environmental data.

2. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 1 is characterized in that: The generating of training content and assessment content according to trainee information includes: Extract the theoretical scores and practical scores of each teaching stage in the student information; generate the theoretical status and practical status of the corresponding teaching stage according to the theoretical scores and practical scores; When the theoretical status is proficient, the advanced theoretical courseware of the corresponding teaching stage is searched in the courseware sub-database in the database as the theoretical training content of the corresponding teaching stage; When the theoretical status is medium, the medium theoretical courseware of the corresponding teaching stage is searched in the courseware sub-database in the database as the theoretical training content of the corresponding teaching stage; When the theoretical state is the basic state, the basic theoretical courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the theoretical training content of the corresponding teaching stage; When the practical operation status is proficient, the advanced practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; When the practical operation status is medium, the medium practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; When the practical operation status is the basic status, the basic practical operation courseware of the corresponding teaching stage is searched in the courseware sub-library in the database as the practical operation training content of the corresponding teaching stage; the training content includes theoretical training content and practical operation training content; Obtain the theoretical and practical status of each teaching stage; The proportion of questions with a proficient theoretical status is set to a high proportion; the proportion of questions with a medium theoretical status is set to a medium proportion; the proportion of questions with a basic theoretical status is set to a low proportion; according to the proportion of questions in each teaching stage, a set number of questions are randomly selected from the theoretical test question sub-database of the database as theoretical assessment content; The proportion of questions with proficient practical status is set to high proportion; the proportion of questions with medium practical status is set to medium proportion; the proportion of questions with basic practical status is set to low proportion; according to the proportion of questions in each teaching stage, a set number of questions are randomly selected from the practical test question sub-database of the database as practical assessment content; the assessment content includes theoretical assessment content and practical assessment content.

3. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 2 is characterized in that: The generating of the theoretical status and the practical status of the corresponding teaching stage according to the theoretical score and the practical score includes: Determine whether the theoretical score is greater than the high theoretical score threshold. If yes, set the theoretical status to the proficient status; if no, when the theoretical score is greater than the medium theoretical score threshold, set the theoretical status to the medium status; otherwise, set the theoretical status to the basic status; Determine whether the practical score is greater than the advanced practical score threshold. If yes, set the practical status to the proficient status; if no, when the theoretical score is greater than the medium practical score threshold, set the practical status to the medium status; otherwise, set the practical status to the basic status.

4. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 1 is characterized in that: The method of constructing a digital twin model of an autoclave according to the autoclave test machine data includes: Extract the autoclave test machine entity data and the autoclave use environment data from the autoclave test machine data; construct the autoclave three-dimensional model according to the autoclave test machine entity data, and construct the autoclave digital twin model according to the autoclave use environment data and the autoclave three-dimensional model; Extract the operable parts in the autoclave test machine data, and establish a mapping relationship between the operable parts of the autoclave test machine and the corresponding operable parts of the autoclave digital twin model.

5. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 1 is characterized in that: The method of using historical practical operation data to train an artificial intelligence model to obtain a prediction result generation model includes: Acquire several groups of historical practical operation data and the corresponding autoclave status labels and instrument data through the database; integrate several practical operation data in each group of practical operation data into a practical operation sequence according to their corresponding time, and obtain the time position of each practical operation data in the practical operation sequence; mark the sequence composed of the practical operation data before the corresponding time position of the corresponding practical operation sequence according to the time order as a partial practical operation sequence of the corresponding practical operation data; Integrate a number of historical practical operation data, and the corresponding partial practical operation sequences and practical operation sequences, and the corresponding autoclave state labels and instrument data into a number of sets of training data and test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model whose input is a partial practical operation sequence and whose output is the corresponding autoclave state label and instrument data is obtained, the instrument data is used as predicted instrument data, the predicted instrument data and the autoclave state label are integrated into predicted result data for output, and finally a prediction result generation model is obtained; wherein the artificial intelligence model includes a BP neural network model and a RBF neural network model.

6. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 1 is characterized in that: The generating of the practical instrument data and the practical feedback signal according to the practical auxiliary model, the practical data and the environmental data includes: Determine whether the practical auxiliary model is a prediction result generation model; If yes, then obtain the practical operation data, integrate the practical operation data from the beginning of this test or training to the present into a partial practical operation sequence in chronological order; input the practical operation sequence into the prediction result generation model to obtain the prediction result data, and use the prediction instrument data in the prediction result data as the practical operation instrument data; when the autoclave status label in the prediction result data is abnormal, set the practical operation feedback signal to abnormal; otherwise, set the practical operation feedback signal to normal; If not, obtain the actual operation data and environmental data, input the actual operation data and environmental data into the digital twin model of the autoclave, obtain the corresponding simulation result data, and use the analog instrument data in the simulation result data as the actual operation instrument data; when the state of the autoclave in the simulation result data is abnormal, set the actual operation feedback signal to abnormal; otherwise, set the actual operation feedback signal to normal; the actual operation feedback signal includes abnormal or normal.

7. The industrial-grade autoclave intelligent training and examination system based on visual-touch collaborative perception according to claim 1 is characterized in that: The alarm module generates an alarm signal according to the actual operation instrument data and the actual operation feedback signal, and issues an alarm according to the alarm signal; The generating of the alarm signal according to the actual operation instrument data and the actual operation feedback signal comprises: When the actual operation feedback signal is abnormal, the instrument value of each instrument in the actual operation instrument data is extracted, and each instrument value exceeding the corresponding set safety range of the instrument is set as an abnormal instrument value, and an alarm signal of the abnormal instrument value is generated; Otherwise, no alarm signal is generated.

Images