Intelligent auxiliary device for simulator training of nuclear power plant license personnel

Abstract

The invention relates to an intelligent auxiliary device for simulator training of nuclear power plant licenses. The intelligent auxiliary device comprises an audio acquisition device, a sound console and a server, the server comprises a voice model module and a voice service module; the audio acquisition device is used for acquiring audio information and respectively sending the audio information to the sound console and the server; the sound console is used for locally storing audio information of each person; the voice model module is used for deploying a voice model; and the voice service module is used for identifying the audio signal based on a voice model deployed by the voice model module, obtaining voice information and keyword identification data of each party in the classroom, and interacting with the auxiliary evaluation system. According to the invention, the operation interaction between the simulator instructor and the simulation platform can be effectively reduced, the teaching energy of the simulator instructor is released, and the training efficiency of the simulator instructor is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power plant simulator training, and more specifically, to an intelligent auxiliary device for nuclear power plant licensee simulator training. Background Art

[0002] The training quality of nuclear power plant licensees is an important talent guarantee for ensuring the safe operation of nuclear power plants, and simulator training is a key link in the training of nuclear power plant licensees. Existing simulator teaching auxiliary devices impose a relatively large burden on the instructor's energy during simulator teaching, and to a certain extent, the simulator training efficiency affects the training quality of licensees. Therefore, a set of intelligent simulator training auxiliary teaching devices needs to be proposed to improve the training efficiency of simulator instructors. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent auxiliary device for nuclear power plant licensee simulator training in view of the problems existing in the prior art.

[0004] The technical solution adopted by the present invention to solve its technical problems is to construct an intelligent auxiliary device for nuclear power plant licensee simulator training, including: an audio acquisition device, a mixer, and a server; the server includes: a voice model module and a voice service module;

[0005] The audio acquisition device is used to collect audio information and send the audio information to the mixer and the server respectively;

[0006] The mixer is used to locally store the audio information of each person;

[0007] The voice model module is used to deploy a voice model;

[0008] The voice service module is used to recognize the audio signal based on the voice model deployed by the voice model module, obtain the voice information and keyword recognition data of all parties in the classroom, and interact with the auxiliary evaluation system.

[0009] In the intelligent auxiliary device for nuclear power plant licensee simulator training of the present invention, it further includes: the auxiliary evaluation system;

[0010] The auxiliary evaluation system communicates with the voice service module, and is used to score based on the voice information of all parties in the classroom and the keyword recognition data, and return a control command to the voice service module.

[0011] In the intelligent auxiliary device for nuclear power plant licensee simulator training of the present invention, the audio acquisition device includes: a wireless microphone, a receiver, an audio distributor, and an encoder;

[0012] The wireless microphone is worn on each person and is used to collect the audio information of each person;

[0013] The receiver communicates with the wireless microphone and is used to receive the audio information and transmit the audio information to the audio distributor;

[0014] The audio distributor is respectively connected to the encoder and the mixer, and is used to send the audio information to the mixer and the encoder respectively;

[0015] The encoder is connected to the server and is used to encode the audio information and then send it to the server.

[0016] In the intelligent auxiliary device for the training of nuclear power plant license personnel simulator of the present invention, the voice service module includes: a data preprocessing unit, a voice recognition service unit, and a voice control service unit;

[0017] The data preprocessing unit is used to preprocess the audio information to obtain preprocessed data;

[0018] The voice recognition service unit is used to perform voice recognition on the preprocessed data to obtain the voice information of all parties in the classroom and record it;

[0019] The voice control service unit is used to perform keyword recognition on the preprocessed data to obtain classroom keywords, and send the classroom keywords to the auxiliary evaluation system.

[0020] In the intelligent auxiliary device for the training of nuclear power plant license personnel simulator of the present invention, the server further includes: a database module;

[0021] The database module is connected to the voice recognition service unit and is used to receive and save the voice information of all parties in the classroom.

[0022] In the intelligent auxiliary device for the training of nuclear power plant license personnel simulator of the present invention, the auxiliary evaluation system includes: a scoring module;

[0023] The scoring module is connected to the database module and is used to obtain the voice information of all parties in the classroom and perform scoring based on the voice information of all parties in the classroom and the scoring rules.

[0024] In the intelligent auxiliary device for the training of nuclear power plant license personnel simulator of the present invention, the auxiliary evaluation system further includes: a course control module;

[0025] The course control module is connected to the voice control service unit and is used to send a control command to the simulation platform based on the classroom keywords, and return a control command to the voice service module.

[0026] In the intelligent auxiliary device for training nuclear power plant license personnel simulators according to the present invention, the data preprocessing unit includes: a data operation sub-module, a data slicing sub-module, an audio truncation sub-module, an audio screening sub-module, a format conversion sub-module, a sampling rate adjustment sub-module, and a noise removal sub-module;

[0027] The data operation sub-module is used to perform operation interpretation processing on the audio information;

[0028] The data slicing sub-module is used to cut and slice the audio files in the audio information;

[0029] The audio truncation sub-module is used to intercept the audio segments in the audio information;

[0030] The audio screening sub-module is used to screen the audio information;

[0031] The format conversion sub-module is used to perform format conversion on the audio information;

[0032] The sampling rate adjustment sub-module is used to uniformly adjust the sampling rate of the audio information;

[0033] The noise removal sub-module is used to remove the silence or background noise in the audio information.

[0034] In the intelligent auxiliary device for training nuclear power plant license personnel simulators according to the present invention, the course control module and the voice control service unit perform data interaction using the http protocol.

[0035] In the intelligent auxiliary device for training nuclear power plant license personnel simulators according to the present invention, the course control module sends control instructions to the simulation platform in the form of natural language.

[0036] Implementing the intelligent auxiliary device for training nuclear power plant license personnel simulators according to the present invention has the following beneficial effects: including: an audio collection device, a mixer, and a server; the server includes: a voice model module and a voice service module; the audio collection device is used to collect audio information and send the audio information to the mixer and the server respectively; the mixer is used to locally store the audio information of each person; the voice model module is used to deploy a voice model; the voice service module is used to identify the audio signal based on the voice model deployed by the voice model module to obtain the voice information and keyword recognition data of all parties in the classroom, and interact with the auxiliary evaluation system. The present invention can effectively reduce the operation interaction between the simulator instructor and the simulation platform, free up the teaching energy of the simulator instructor, and improve the training efficiency of the simulator instructor. Description of the Drawings

[0037] The present invention will be further described below in conjunction with the accompanying drawings and embodiments. In the accompanying drawings:

[0038] Figure 1 is a schematic structural diagram of an intelligent auxiliary device for training nuclear power plant license personnel simulator provided by the present invention;

[0039] Figure 2 is a functional block diagram of a server provided by the present invention;

[0040] Figure 3 is a flowchart of an embodiment of intelligent personnel simulator training evaluation provided by the present invention;

[0041] Figure 4 is an operation interface diagram for querying business scenario information provided by the present invention;

[0042] Figure 5 is a diagram of a scene device and a score display interface provided by the present invention;

[0043] Figure 6 is a business process flowchart for implementing simulator teaching / evaluation provided by the present invention;

[0044] Figure 7 is a flowchart of intelligent teaching evaluation provided by the present invention. Detailed Embodiments

[0045] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] The present invention provides an intelligent auxiliary device for training nuclear power plant license personnel simulator, which applies voice recognition and voice interaction technologies to the training of nuclear power plant license personnel simulator, realizes reducing the operation interaction between simulator instructors and the simulation platform, releasing the teaching energy of simulator instructors, and effectively improving the training efficiency of simulator instructors. Among them, the simulation is a full-scope simulation platform for nuclear power plants, namely 3KeyMaster.

[0047] Specifically, as Figure 1As shown in the figure, the intelligent auxiliary device for the licensee simulator training of the nuclear power plant includes: an audio acquisition device 100, a mixing console 200, and a server 300. Among them, the server 300 includes: a voice model module 301, a voice service module 302, and an auxiliary evaluation system 304. In the embodiment of the present invention, the audio acquisition device 100 is used to acquire audio information and send the audio information to the mixing console 200 and the server 300 respectively; the mixing console 200 is used to locally store the audio information of each person; the voice model module 301 is used to deploy a voice model; the voice service module 302 is used to identify the audio signal based on the voice model deployed by the voice model module 301 to obtain the voice information of all parties in the classroom and keyword recognition data, and interact with the auxiliary evaluation system 304. The auxiliary evaluation system 304 communicates with the voice service module 302, and is used to score based on the voice information of all parties in the classroom and keyword recognition data, and return a control command to the voice service module 302.

[0048] In some embodiments, as Figure 1 shown in the figure, the audio acquisition device 100 includes: a wireless microphone 101, a receiver 102, an audio distributor 103, and an encoder 105; the wireless microphone 101 is worn on each person to acquire the audio information of each person. Among them, the receiver 102 communicates with the wireless microphone 101 and is used to receive the audio information and transmit the audio information to the audio distributor 103; the audio distributor 103 is respectively connected to the encoder 105 and the mixing console 200, and is used to send the audio information to the mixing console 200 and the encoder 105 respectively. The encoder 105 is connected to the server 300 and is used to encode the audio information and then send it to the server 300.

[0049] Specifically, in order to reduce the wearing load of the simulator instructor (playing the roles of various professional positions in the nuclear power plant), the primary loop operator, the secondary loop operator, the unit leader, the shift supervisor, and the nuclear safety technical advisor, in the embodiment of the present invention, the wireless microphone 101 is of a lapel type and is respectively worn on the chests of each person. The audio information of each person is acquired by the wireless microphone 101 worn on the chest of each person, and the acquired audio information is sent to the receiver 102 in a wireless manner. Then, the receiver 102 is connected to the audio distributor 103 through a cable, and the receiver 102 transmits the received audio information to the audio distributor 103. Then, the audio distributor 103 sends the received audio information of each person to the mixing console 200 through a cable in one way and to the encoder 105 in the other way. Among them, the mixing console 200 is used to locally store the audio information of each person. The encoder 105 encodes the audio information of each person and then transmits it to the server 300 through a cable. In the embodiment of the present invention, the server 300 can be set in the workstation of the nuclear power plant.

[0050] In an embodiment of the present invention, a high-performance graphics card is configured in the server 300, and at the same time, a speech recognition module, various speech services, an intermediate module for communicating with an external system, etc. are deployed. Through the intelligent auxiliary device for nuclear power plant licensee simulator training, all the language expressions of the teaching scenario roles during the full-range teaching can be real-time converted into text information, which is convenient for teaching review and analysis. Optionally, the server 300 can adopt a 4090 computing card server 300.

[0051] In an embodiment of the present invention, the speech model module 301 is mainly used to deploy models. Among them, the models deployed by the speech model module 301 can be selected for single-model deployment or multi-model deployment according to specific situations. In single-model deployment, all speech streams share one model, and in multi-model deployment, each speech stream is configured with one model. For example, in a project where the speech traffic of each personnel role is relatively small and the concurrency is not high, single-model deployment can be preferentially used. Among them, Figure 2 The figure shows an example of multi-model deployment.

[0052] In an embodiment of the present invention, the speech service module 302 is mainly used to receive speech streams, the control streams of the auxiliary evaluation system 304, and process the received speech streams and the results after recognition, and directly interact with the auxiliary evaluation system 304.

[0053] Optionally, in an embodiment of the present invention, the speech service module 302 includes: a data preprocessing unit, a speech recognition service unit, and a speech control service unit. Among them, the data preprocessing unit is used to preprocess audio information to obtain preprocessed data; the speech recognition service unit is used to perform speech recognition on the preprocessed data to obtain the speech information of all parties in the classroom and record it; the speech control service unit is used to perform keyword recognition on the preprocessed data to obtain classroom keywords, and send the classroom keywords to the auxiliary evaluation system 304.

[0054] Optionally, in an embodiment of the present invention, the data preprocessing unit includes: a data operation sub-module, a data sharding sub-module, an audio truncation sub-module, an audio screening sub-module, a format conversion sub-module, a sampling rate adjustment sub-module, and a noise removal sub-module; the data operation sub-module is used to perform operation interpretation processing on audio information; the data sharding sub-module is used to cut and shard the audio files in the audio information; the audio truncation sub-module is used to intercept the audio segments in the audio information; the audio screening sub-module is used to screen the audio information; the format conversion sub-module is used to perform format conversion on the audio information; the sampling rate adjustment sub-module is used to uniformly adjust the sampling rate of the audio information; the noise removal sub-module is used to remove the silence or background noise in the audio information.

[0055] Specifically, in the embodiments of the present invention, the speech recognition service unit is mainly used to recognize the speech stream to obtain the speech information of all parties in the classroom and record the speech information of all parties in the classroom. At the same time, as Figure 2 shown, the speech recognition service unit is also connected to the auxiliary evaluation system 304 through a database. After recognizing the speech, it can be directly written into the database for the auxiliary evaluation system 304 to score. Moreover, the speech recognition service unit also retains the wav format records of each audio source.

[0056] Furthermore, in the embodiments of the present invention, before performing the speech recognition service, data preprocessing is required. By preprocessing the data through the data preprocessing unit, the recognition accuracy and efficiency can be improved. Among them, the specific preprocessing operations are as follows:

[0057]

[0058] By performing the above operations to preprocess the audio data, the accuracy of speech recognition can be significantly improved. Among them, the data preprocessing operation is in the form of an http service, which docks with the audio provider (i.e., the encoder 105) to receive the audio file of the encoder 105. This service is implemented in the form of a python server. After performing the speech recognition service, the deployed pytorch version of the whisper model is used for online service in the form of http. Among them, the steps of the http speech recognition service are as follows: Load the pre-trained whisper model into the server 300 and perform necessary optimizations (for example, use mixed precision or tensor decomposition technology for optimization to improve the inference speed and reduce resource consumption). Then use a python framework such as flask to receive the preprocessed audio data. Finally, return the recognition result in the form of http. Among them, the speech recognition data table is as follows:

[0059]

[0060] Among them, class_id is used as a partition field to improve the query efficiency, and user_id is used to distinguish audio sources, which can be configured in the configuration file. Various times are used to determine the absolute and relative times of the audio. For the final field, whisper itself does not support streaming recognition tasks, so the quasi-real-time recognition method used in this project is real-time segmentation. This results in the recognition result of the last segment being continuously updated until a new segment is started by the segmentation module. To ensure the real-time performance of the recognition, the continuously updated recognition results can be retained and marked with the final field. There will be at most one (none if considered silent) final recognition result for the speech at the same time, and possibly multiple temporary recognition results.

[0061] Among them, the accuracy requirement for speech recognition is as follows: adopting the mainstream evaluation index of current speech recognition, the word error rate (WER), ensuring that the speech recognition device can learn nuclear power plant professional terms, with the word error rate of natural language not higher than 85%, and the word error rate corresponding to nuclear power plant professional terms not higher than 70%, realizing an important auxiliary tool for AI technology in the full-scope simulator teaching scenario of nuclear power plants. For the safety requirement of AI technology: due to the unique safety requirements of nuclear power plants, while ensuring functional requirements, offline large pre-research model technology is also required to ensure the physical isolation of nuclear power plant data from external networks and ensure information security.

[0062] In the embodiment of the present invention, the voice control service unit is mainly used to, during teaching and training, realize that when the simulator instructor inserts in the simulator teaching scenario, identify keywords of the preprocessed data, obtain classroom keywords, and send the classroom keywords to the auxiliary evaluation system 304. The auxiliary evaluation system 304 sends control instructions to 3KeyMaster in a natural language manner and converts them into specific control instructions of 3KeyMaster, so as to reduce the operation interaction between the simulator instructor and 3KeyMaster and release the teaching energy of the simulator instructor. Among them, voice control is to use voice to start or end the corresponding stage of the scenario during the course. In the setting of the voice input channel, it can be set to only detect the control of the instructor's channel. Through the auxiliary evaluation system 304, the following functions can be realized:

[0063] (1) Realize the intelligent management of the simulator training scenario of nuclear power plant operators and the standard:

[0064] The present invention can, according to the sequence, time interval and logical conditions of teaching scenarios set in teaching training lesson plans of different editions of simulators, combined with the simulator teaching plan, automatically identify and trigger teaching scenarios after the teaching scenarios start, realize the automatic management of the simulator teaching process, reduce the energy of the simulator instructor in managing teaching scenarios during the teaching process, and avoid mis-triggering teaching scenarios and causing distortion of the simulation scenario.

[0065] (2) Intelligent evaluation of key points for teaching observation:

[0066] The current simulator teaching completely relies on simulator instructors to collect a large amount of system equipment and parameters to evaluate the performance of trainees, which has certain limitations. At the same time, relevant information needs to be recorded in real time for review after the course, which greatly involves the observation energy of simulator instructors. The present invention can, according to rules set in advance for different scenarios, automatically capture the state of relevant system equipment and the change process of parameters in the scenario, automatically generate operation logs of trainees, automatically analyze the control effect of the crew, and automatically generate and archive the evaluation performance of trainees in chronological order. While reducing the observation load of simulator instructors, it improves the comprehensiveness and accuracy of the observation of simulator instructors' teaching.

[0067] (3) Assist in observing the specific behavior performance of trainees in simulator teaching (such as three-stage communication, etc.):

[0068] The present invention converts the voice data of trainees in the teaching scenario into text information through voice recognition, and then retrieves and analyzes the text information through information integration technology to determine whether trainees implement specific behavior norms according to management requirements, such as "three-stage communication", "self-voting", and "guardianship system", etc. It realizes the observation and intelligent evaluation of trainees' behavior performance in the teaching scenario with minimal investment.

[0069] (4) Assist in carrying out intelligent analysis of ability status and improvement of teaching scenarios:

[0070] The present invention, through the data generated during the simulator teaching process, combines with the scoring situation of the operator ability evaluation standard, intelligently analyzes the ability status of the operator from individual to organization, identifies weak ability items, so as to assist relevant personnel in nuclear power plant training business to carry out targeted ability improvement activities. At the same time, it gives the performance of trainees in different scenarios during training, providing a reference for the optimization of training strategies.

[0071] Optionally, in the embodiment of the present invention, the control detection words used for control detection are general words, rather than selected terms or course numbers. Among them, the current control detection words are as follows:

[0072] Startup scenario stage: Start the Xth stage of the Xth scenario;

[0073] End scenario stage: End the Xth stage of the Xth scenario.

[0074] It should be noted that in the embodiment of the present invention, the control detection words can be changed, but the detection rates of different control detection words are different.

[0075] Furthermore, such as Figure 2As shown, the server 300 further includes: a database module 303; the database module 303 is connected to the speech recognition service unit and is used to receive the speech information of all parties in the classroom and save it. Among them, the database can use a mysql database. Of course, it can also use but is not limited to: databases such as redis database, mongoDB database, kafka database, etc.

[0076] As Figure 2 shown, in the embodiment of the present invention, the auxiliary evaluation system 304 includes: a scoring module; the scoring module is connected to the database module 303 and is used to obtain the speech information of all parties in the classroom and perform scoring based on the speech information of all parties in the classroom and the scoring rules.

[0077] Further, as Figure 2 shown, the auxiliary evaluation system 304 further includes: a course control module; the course control module is connected to the speech control service unit and is used to send a control instruction to the simulation platform based on the classroom keywords and return a control command to the speech service module 302. Among them, Figure 2 the other modules in

[0078] may include but are not limited to: a lesson preparation module, a scoring formula module, a performance analysis module, an interface module, etc.

[0079] {

[0080] "class_id":"xxx_xxx_xxx",

[0081] "timestamp":"1722243648000",

[0082] "action":"start",

[0083] "scene":1,

[0084] "step":2

[0085] }

[0086] The trigger keyword to close the scene, accepting json:

[0087] {

[0088] "class_id":"xxx_xxx_xxx",

[0089] "timestamp":"1722243648000",

[0090] "action":"end",

[0091] "scene":1,

[0092] "step":2

[0093] }

[0094] In the embodiments of the present invention, in order to realize the interaction between natural language and the 3KeyMaster system, it is necessary to rely on the interface relationship between the auxiliary evaluation system 304 and the 3KeyMaster system to ensure that the natural language containing nuclear power plant professional terms can be seamlessly docked with the virtual points of the system equipment in the 3KeyMaster system, and at the same time, the feedback information of the operation instructions of the 3KeyMaster is fed back to the simulator instructor by voice.

[0095] Specifically, through the interface module of the auxiliary evaluation system 304, the intelligent auxiliary device for the training of nuclear power plant license personnel, the auxiliary evaluation system 304 and the 3KeyMaster system can be connected to form a completely independent and applicable intelligent operator simulator training and evaluation method for nuclear power plants, thereby optimizing the traditional method of relying on simulation instructors to observe and evaluate the performance of students based on their own knowledge and experience during the simulation teaching process, and contributing to the cultivation of nuclear power plant license personnel. Among them, the specific evaluation process is as Figure 3 shown. The present invention is completely based on the training courses and assessment systems of nuclear power plant operators, and systematically manages the training system of operators. The relevant course information is obtained from the training management system by a barcode scanner, providing the function of further maintaining the training project information. The auxiliary evaluation system 304 connects the relationships between training, courses, and business scenarios, making the relationships of the data of each module clear and the structure clear, simplifying the operation of users, and improving the unity and reusability of the data of each module. Different from the conventional teaching field, the smallest unit scene in the simulator scene library created by the present invention directly includes the teaching scene content undertaken by national regulations and industry standards. At the same time, for the subsequent intelligent scene implementation, the unit scene needs to be matched with the command instructions in the 3KeyMaster to prepare for inserting the corresponding scene by voice or manually, including information such as simulator trigger roll call, implementation method, delay, and whether it is expected to protect.

[0096] At the same time, for each scenario, according to the needs of simulator teaching, it can be divided into multiple stages, so that the lesson plan designer can select the corresponding stage for teaching and evaluation according to the requirements. Each stage should include its corresponding evaluation observation items. Each observation and evaluation item is also associated with the points in 3KeyMaster and DCS (S-VDU), and generates the machine language recognizable by the corresponding simulator system according to a specific formula. At the same time, it can return the evaluation content of whether each observation and evaluation item is successful and its corresponding score. When a protection signal is expected to exist, it can also be recognized in advance. The following takes the SGTR(SI) scenario as an example for demonstration.

[0097] Scenario name: SGTR of Steam Generator 1 and resulting in the triggering of ECCS

[0098] Implementation name: mfRCP010;

[0099] Implementation method: 20m 3 / h;

[0100] Delay: 20;

[0101] Observation and evaluation items in Phase 1

[0102]

[0103]

[0104] Among them, the simulator teaching / assessment / special practice scenarios are based on the auxiliary observation system and the smallest scenario unit in the scenario library. According to the teaching and assessment objectives, they are combined in sequence to form the simulator teaching / assessment / special practice scenarios, realizing a strong association between the lesson plan and the simulator. The relevant operation interfaces and display interfaces are as Figure 4 and Figure 5 shown. Finally, according to the teaching objectives, the triggering time sequence of the scenario is preset in advance, starting from the first scenario unit as the time sequence starting point, providing basic data for the implementation of automated teaching.

[0105] In some embodiments, the simulator teaching / assessment is implemented as follows:

[0106] Before the implementation of simulator teaching / assessment, the interfaces with the virtual DCS system, 3KeyMaster system, operation prompt system, and audio system will be verified. Select the training course for the day, enter the on-site evaluation interface, collect real-time simulator operation data and voice command texts, calculate according to the evaluation rules, and retain the calculation results of the evaluation rules. After the training operation is completed, the system will automatically calculate the scores of all evaluation items. If there is no score result, it will prompt the instructor to check and verify, and perform manual scoring. After the instructor checks and it is completely correct, save the evaluation results of this operation training. The system provides function operations for evaluation start, automatic calculation of scores, manual scoring, and evaluation end. Its business process is as Figure 6 shown.

[0107] In some embodiments, the implementation of inserting teaching / assessment scenarios by voice or manually is as follows:

[0108] The present invention can realize inserting corresponding teaching scenarios by voice or manually in the auxiliary observation system. Through the interaction with the "intelligent auxiliary device for simulator training", and the point position configuration of natural language with the virtual DCS system, 3KeyMaster system, operation prompt system, and audio system, the intelligent interaction between human natural language and machine language can be realized, and thus the cumbersome human-machine interaction process between the simulator instructor and the simulator system can be reduced or even avoided.

[0109] At the same time, according to the sequence, time interval, and logical conditions of the teaching scenarios set in the teaching plan, combined with the simulator teaching plan, after the teaching scenario starts, it automatically identifies and triggers the teaching scenario, realizes the automatic management of the simulator teaching process, reduces the energy of the simulator instructor in managing the teaching scenario during the teaching process, and avoids mis-triggering the teaching scenario, resulting in distortion of the simulation scenario. Thus, the automatic triggering of the teaching scenario is realized. According to the designed scenario triggering logic, according to certain triggering conditions, it reminds the simulator instructor whether to trigger the scenario in a voice manner. After obtaining voice or key confirmation, it automatically triggers the scenario in 3KM, realizing the "autopilot" function of simulator teaching.

[0110] Taking the voice activation of the XX-year MS fourth unit scenario as an example, present its command operation flow:

[0111] ① Load the simulator IC:

[0112] Simulator instructor -> "Intelligent auxiliary device for simulator training": Start the MS fourth-day scenario;

[0113] "Intelligent auxiliary device for simulator training" -> "Auxiliary observation system": Start the MS fourth-day scenario;

[0114] "Auxiliary observation system" -> 3KeyMaster: Start the MS fourth-day scenario;

[0115] 3KeyMaster: Load IC224 and run.

[0116] ② Insert the simulator scenario:

[0117] Simulator instructor -> "Intelligent Assistant Device for Simulator Training": Insert the first scenario;

[0118] "Intelligent Assistant Device for Simulator Training" -> "Auxiliary Observation System": Insert the first scenario;

[0119] "Auxiliary Observation System" -> 3KeyMaster: Insert the internal leakage of GCT121VV;

[0120] 3KeyMaster: Execute List: 2025MS4.1 - GCT121VV (PID → GCT121VV → internal leakage F = 0.5RT = 300s);

[0121] After 30 minutes: 3KeyMaster: Execute List: 2025MS4.2 - RCP003PO (3KM → F4 → RCV040MD → RCV880 / 883VP → position = 0);

[0122] After 30 minutes: 3KeyMaster: Execute List: 2025MS4.3 - RRA021VP (PID - RRA021VP - internal leakage - F = 2×10 - 5 (1kg / s), RT = 50s);

[0123] After 2 minutes: 3KeyMaster: Execute 2025MS4.4 - RRA115VP (PID - RRA115VP - internal leakage - F = 0.001 (0.09kg / s) RT = 100);

[0124] After 28 minutes: 3KeyMaster: Execute List: 2025MS4.4 - RCV (mfRCV21 F = 0.2RT = 60);

[0125] 3KeyMaster: Execute List: 2025MS4.4 - RCP355VP (PID - RCP355VP F = 0RT = 100);

[0126] 3KeyMaster: Execute List: 2025MS4.4 - PTR022VB OPEN (PID - PTR022VB F = 1 RT = 50);

[0127] 3 KeyMaster: Execute List: 2025MS4.4 - PTR022VB CLOSE (PID - PTR022VB F = 0 RT = 50);

[0128] After 40 minutes: 3 KeyMaster: Execute List: 2025MS4.5 - RRA115VP Expand (PID RRA115VP internal leakage F = 1 RT = 100);

[0129] After 30 minutes: 3 KeyMaster: Execute List - 2025MS4.6 → RCV001PO (PID - RCV132 / 133MV - zero drift - F = 0.45 / 0.33);

[0130] After 20 minutes 3 KeyMaster: Execute List - 2025MS4.7 → LOC.

[0131] In some embodiments, the implementation of the cooperation instructions is as follows:

[0132] During the teaching process, when the simulator instructor needs to play the role of professionals and on - site operators in the actual production process and cooperate with the operators in the main control room to carry out corresponding on - site work, the simulator instructor can send voice cooperation instructions through the "intelligent auxiliary device for simulator training". By realizing the voice function, the hands of the simulator instructor can be liberated. The voice inserts the on - site operator and professional's instructions for equipment and on - site cooperation in the simulator teaching scenario, greatly reducing the cooperation operations of the simulator instructor and enabling them to focus on the observation and evaluation of simulator teaching.

[0133] The following is the instruction operation flow:

[0134] Main control operator -> Simulator instructor (role of on - site personnel): On - site cooperation to execute RFLL006;

[0135] Simulator instructor (role of on - site personnel) -> "Intelligent auxiliary device for simulator training": Execute RFLL006;

[0136] "Intelligent auxiliary device for simulator training" -> "Auxiliary observation system": Execute RFLL006;

[0137] "Auxiliary observation system" -> 3 KeyMaster: Execute RFLL006.

[0138] 3 KeyMaster: Execute the RFLL006 instruction:

[0139] +1 ICM aov 8TER049VE t:1 "Close 9TER 049VE (QA201)";

[0140] +1IRF rfrfTER04 f:0 “Turn off 9TER 051VE”;

[0141] +1IRF rfSEL14 f:0 “Turn off 9SEL 044VE”;

[0142] +1IRF rfTER10 f:0 “Turn off 9TER 036VE”;

[0143] +1IRF rfTER23 f:1 “Turn off 9TER037VE”;

[0144] +1IRF rITER12 f:1 “Turn on 9TER 093VE”;

[0145] +1IRF rfTER08 f:1 “Turn on 9TER 038VE”;

[0146] +1IRF IFAPG14 f:1 “Turn on APG 007VL”;

[0147] +1IRF rfAPG15 f:1 “Turn on APG008VL”;

[0148] +1IRF rfAPG16 f:1 “Turn on APG009VL”;

[0149] +1IRF rfAPG09 f:1 “Turn on APG 018VL”;

[0150] +1IRF rfAPG11 r:20f:0 “Turn off APG 041VL”;

[0151] +1IRF IFAPG08 r:20f:0 “Turn off APG 017VL”;

[0152] +1IRE fAPG10 I:20f:0 “Turn off APG 040VL”.

[0153] In some embodiments, for intelligent teaching evaluation:

[0154] When the aforementioned minimum unit scenario is executed, the corresponding observation and evaluation items of the scenario can be automatically called up, and the corresponding parameters and device status are captured from the virtual DCS system, 3KeyMaster system, operation prompt system, and audio system (i.e., voice service module 302) for observation and evaluation, and an observation and evaluation return statement is output, which is used as a factual record of the trainee's behavioral performance and a corresponding score is generated. Thus, according to the observation elements of the simulator scenario, the evaluation of the crew control effect of the licensed personnel in the teaching scenario is intelligently generated, and the evaluation comment record and the change trend of key parameters are automatically generated in real time, the precise matching of specific behavior specifications, the analysis of the protection signals of the scenario, and the teaching observation efficiency of the simulator is improved while enhancing the comprehensiveness and accuracy of the teaching observation. The process is as Figure 7 shown.

[0155] Taking the aforementioned SGTR scenario as an example, for the observation item "correctly execute KOO001KG placed in the ON position", during the observation of this scenario, if the simulator captures the instruction that the trainee places KOO001KG in the ON position, at this time the observation item is marked as completed, and at the same time a score of 5 points will be output and the following description of the behavioral fact will be returned: "00:12:32 RO1 correctly executes the confirmation operation of KOO001KG".

[0156]

[0157] Among them, for the full-information database management and weak-point analysis: All the data generated by the above-mentioned trainees during the simulator training courses / simulator assessments / simulator special exercises will be aggregated in the "Auxiliary Observation System" database to form a database of the capabilities of nuclear power plant operators, which serves as the input for the statistical analysis of the trainees' ability status and scenario performance. For the course scenario statistics: Through the course scenario statistics, the distribution of scenarios under different versions in the simulator teaching / assessment / special exercises, the usage frequency of different scenarios, and the scenario usage compliance can be intuitively viewed, so as to intuitively present the instructors' preferences for scenario usage, in order to better serve the upgrade of teaching plans. For the scenario usage statistics: The scenario usage statistics can complete the frequency ranking of all the scenarios used, and then analyze the trainees' focus on scenario practice during the simulator exercises, providing a reference for subsequent exercises. For the scenario performance statistics: For the scenario performance, based on the scoring of each observation and evaluation item for each scenario, the performance differences of current nuclear power plant license holders in different scenarios can be statistically analyzed, and then more targeted training can be carried out, laying a foundation for the optimization and adjustment of training strategies. For the course performance analysis: It is used to analyze the overall performance of trainees at different levels of personnel, different operating values, different departments, and the company level, intuitively informing the management of the existing ability advantages and weaknesses of the current company's license holder team, providing a direction for the cultivation of the company's operator capabilities. At the same time, it can enable individual license holders to clearly identify their own main ability weaknesses and point out the direction for their own improvement in subsequent work and study. For the unexpected protection action analysis: The number of different protection signal actions and various protection situations triggered by different personnel in different scenarios can be counted, which will serve as an important input for the personal ability portrait of operators.

[0158] In some embodiments, the interface module of the auxiliary evaluation system 304 includes, but is not limited to: DCS interface, 3Keymaster interface, operation prompt system interface, simulator training intelligent auxiliary device interface, training system interface, etc.

[0159] The DCS interface realizes the docking of operation-related information of the virtual DCS system. The operation-related information provided by the virtual DCS system includes, but is not limited to, data items, attribute and data value lists, training operation-related real-time data, etc. These operation-related data are accessed through the DCS interface to this intelligent auxiliary device, and the transmission frequency is 1 time / second. Among them, the virtual DCS system and the intelligent auxiliary device use the message queue method for the transmission of interface raw data. The interface raw data is not saved locally. After filtering the raw data through the evaluation rules of the auxiliary evaluation system 304, it is saved locally. Due to the large amount of data, a storage strategy generally needs to be established. The relevant description of the DCS interface is shown in Table 1 below.

[0160] Table 1

[0161]

[0162] The 3Keymaster interface implements the operation-related information for docking with the 3KeyMaster system, and data transfer is achieved between this system and the 3KeyMaster system through this 3Keymaster interface. Specifically, the 3KeyMaster system sends operation data to the auxiliary evaluation system 304 every second, and the auxiliary evaluation system 304 sends instructions for in-situ faults, special faults, and general faults to the 3KeyMaster system. Among them, the relevant descriptions of the 3Keymaster interface are shown in Table 2 below.

[0163] Table 2

[0164]

[0165]

[0166] The operation prompt system interface implements the relevant information for docking with the operation prompt system, and information sharing is achieved between this system and the operation prompt system through this operation prompt system interface. Specifically, when the operator operates in the 3KeyMaster system and the virtual DCS system, the operation prompt system will give corresponding prompt information according to the business scenario; the system provides function operations such as data acquisition and data cleaning for the interface. Among them, the relevant descriptions of the operation prompt system interface are shown in Table 3 below.

[0167] Table 3

[0168]

[0169] The simulator training intelligent auxiliary device interface is responsible for receiving the text information recognized by the voice service module 302, sending instructions to the voice control module to send voice instructions to the 3Keymaster, such as in systems like loading scenarios Scenarios list, inserting faults, and in-situ implementation, etc., and is responsible for displaying the text information recognized by the voice. Among them, the relevant descriptions of the simulator training intelligent auxiliary device interface are shown in Table 4 below.

[0170] Table 4

[0171]

[0172]

[0173] The training system interface is implemented by using a barcode scanner. Specifically, the data entry function is achieved by scanning the QR code. This system obtains data from the training management system in the office intranet, including training course information and training task information, and transmits it in a unified fixed format. After the data is received, it is saved in the evaluation system. Among them, the relevant descriptions of the training system interface are shown in Table 5 below.

[0174] Table 5

[0175]

[0176]

[0177] It should be noted that the requirements for the intelligent auxiliary device for nuclear power plant licensee simulator training in the present invention in the full - scope simulator layout are: deployment in the same network segment of the local area network.

[0178] Through the intelligent auxiliary device for nuclear power plant licensee simulator training in the present invention, when the simulator instructor inserts into the simulator teaching scenario, he can send control instructions to 3KeyMaster in a natural - language manner and convert them into specific control instructions of 3KeyMaster, so as to reduce the operation interaction between the simulator instructor and 3KeyMaster and release the teaching energy of the simulator instructor.

[0179] Specifically, the present invention can achieve the automatic triggering of teaching scenarios. That is, according to the designed scenario - triggering logic and certain triggering conditions, the simulator instructor is reminded by voice whether to trigger the scenario. After obtaining voice or key confirmation, the scenario is automatically triggered in 3KeyMaster, realizing the "autopilot" function of simulator teaching.

[0180] Through the implementation of the voice function in the present invention, the hands of the simulator instructor can be liberated. Specifically, the voice inserts the instructions of on - site operators and professionals for equipment and on - site cooperation in the simulator teaching scenario, greatly reducing the cooperative operations of the simulator instructor and enabling him to focus on the observation and evaluation of simulator teaching.

[0181] The present invention can achieve intelligent teaching evaluation. Specifically, according to the observation elements of the simulator scenario, it intelligently generates the evaluation of the control effect of licensees in the teaching scenario on the unit, and automatically generates the evaluation comment record and the change trend of key parameters in real - time, accurately matches specific behavior specifications, and analyzes the protection signals of the scenario, improving the observation efficiency of simulator teaching while enhancing the comprehensiveness and accuracy of teaching observation.

[0182] The present invention can also achieve full - information database management and weak - point analysis. Specifically, the present invention constructs databases of unit control and behavior specifications of different trainees in different scenarios, and then intelligently locates the ability weak - points of individual operators and organizations, providing a reference for the ability - improvement plan. At the same time, by analyzing the data performance in different scenarios, it is possible to conduct hierarchical training for different types of simulator scenarios and optimize the training strategy.

[0183] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the similarities and common parts among the various embodiments, reference can be made to each other. For the apparatuses disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0184] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0185] The steps of the methods or algorithms described in combination with the embodiments disclosed herein can be directly implemented by hardware, software modules executed by a processor, or a combination of the two. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0186] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They do not limit the protection scope of the present invention. Any equivalent changes and modifications made to the scope of the claims of the present invention shall fall within the scope covered by the claims of the present invention.

Claims

1. An intelligent auxiliary device for the training of licensed personnel simulators in nuclear power plants, characterized in that, Including: An audio acquisition device, a mixing console, and a server; the server includes: a voice model module and a voice service module; The audio acquisition device is used to acquire audio information and send the audio information to the mixing console and the server respectively; The mixing console is used to locally save the audio information of each person; The voice model module is used to deploy a voice model; The voice service module is used to identify the audio signal based on the voice model deployed by the voice model module, obtain the voice information of all parties in the classroom and keyword recognition data, and interact with the auxiliary evaluation system.

2. The intelligent auxiliary device for training of nuclear power plant license personnel simulators according to claim 1, wherein It also includes: The auxiliary evaluation system; The auxiliary evaluation system communicates with the voice service module, and is used to score based on the voice information of all parties in the classroom and the keyword recognition data, and return a control command to the voice service module.

3. The intelligent auxiliary device for nuclear power plant licensee simulator training according to claim 1, wherein The audio acquisition device includes: a wireless microphone, a receiver, an audio distributor, and an encoder; The wireless microphone is worn on each person and is used to acquire the audio information of each person; The receiver communicates with the wireless microphone and is used to receive the audio information and transmit the audio information to the audio distributor; The audio distributor is respectively connected to the encoder and the mixing console, and is used to send the audio information to the mixing console and the encoder respectively; The encoder is connected to the server and is used to encode the audio information and then send it to the server.

4. The intelligent auxiliary device for training nuclear power plant license personnel simulator according to claim 2, characterized in that, The voice service module includes: a data preprocessing unit, a voice recognition service unit, and a voice control service unit; The data preprocessing unit is used to preprocess the audio information to obtain preprocessed data; The voice recognition service unit is used to perform voice recognition on the preprocessed data to obtain and record the voice information of all parties in the classroom; The voice control service unit is used to perform keyword recognition on the preprocessed data to obtain classroom keywords, and send the classroom keywords to the auxiliary evaluation system.

5. The intelligent auxiliary device for training nuclear power plant license personnel simulator according to claim 4, characterized in that, The server also includes: a database module; The database module is connected to the voice recognition service unit and is used to receive and save the voice information of all parties in the classroom.

6. The intelligent auxiliary device for nuclear power plant licensee simulator training according to claim 5, characterized in that, The auxiliary evaluation system includes: a scoring module; The scoring module is connected to the database module and is used to obtain the voice information of all parties in the classroom and score based on the voice information of all parties in the classroom and the scoring rules.

7. The intelligent auxiliary device for nuclear power plant licensee simulator training according to claim 4, characterized in that, The auxiliary evaluation system also includes: a course control module; The course control module is connected to the voice control service unit and is used to send a control instruction to the simulation platform based on the classroom keywords, and return a control command to the voice service module.

8. The intelligent auxiliary device for nuclear power plant licensee simulator training according to claim 4, wherein The data preprocessing unit includes: a data operation sub-module, a data slicing sub-module, an audio truncation sub-module, an audio screening sub-module, a format conversion sub-module, a sampling rate adjustment sub-module, and a silence removal sub-module; The data operation sub-module is used to perform operation interpretation processing on the audio information; The data slicing sub-module is used to cut and slice the audio files in the audio information; The audio truncation sub-module is used to intercept audio segments in the audio information; The audio screening sub-module is used to screen the audio information; The format conversion sub-module is used to perform format conversion on the audio information; The sampling rate adjustment sub-module is used to uniformly adjust the sampling rate of the audio information; The noise removal sub-module is used to remove silence or background noise in the audio information.

9. The intelligent auxiliary device for training nuclear power plant license personnel simulator according to claim 7, characterized in that, The course control module and the voice control service unit perform data interaction using the http protocol.

10. The intelligent auxiliary device for training nuclear power plant license personnel simulator according to claim 7, characterized in that, The course control module sends control instructions to the simulation platform in the form of natural language.

Images