Nuclear power welding specification information fusion management system based on artificial intelligence

Abstract

The invention provides a nuclear power welding specification information fusion management system based on artificial intelligence, and relates to the field of nuclear power digital data processing, the nuclear power welding specification information fusion management system comprises a knowledge base module, a data fusion module, an AI reasoning module, a user management module and a mobile terminal module, the data fusion module is used for integrating different welding processes and outputting a welding scheme, the AI reasoning module is used for conducting reasoning analysis on the processes in the welding scheme and generating process parameters, and the user management module is used for managing qualification information and operation records of welders. The mobile terminal module is used for inquiring and feeding back field information; the system can intelligently analyze a complex nuclear power welding task and generate a detailed scheme, and assign a corresponding welder to work, so that the welding requirement is met.

Description

Technical Field

[0001] The invention relates to the field of nuclear power digital data processing, and in particular to an artificial intelligence-based nuclear power welding specification information fusion management system. Background Art

[0002] In nuclear power and other high-demand industrial fields, the quality and reliability of welding technology are directly related to the overall safety and durability of the project. Welding information management covers the entire process from welding materials, process parameters to quality inspection, involving welding standards, welder qualifications, process design, quality control and risk assessment. Since the welding process involves a large number of parameters, steps and quality control points, the complexity of various information is high, and the existing manual management and traditional information management methods often cannot ensure the real-time update of welding data, the standardization of processes and the quality traceability of the entire process.

[0003] The foregoing discussion of the background art is intended only to facilitate an understanding of the present invention. This discussion does not acknowledge or admit that any of the material referred to is part of the common general knowledge.

[0004] Now many welding management systems have been developed. After a lot of searching and reference, it is found that the existing welding management system is such as the system disclosed by the publication number CN103426066A. These systems generally include A. Knowledge representation of each sub-item of nuclear power welder qualification and establishment of corresponding static database; B. According to different nuclear power welder management standards and specifications, on the basis of knowledge representation, establish logical replacement relationship between sub-items and within sub-items; C. Use the rule engine to solidify and store the logical replacement relationship of project code sub-items and form a special module; D. In the management process of nuclear power welder qualification, through the recording and storage of data, the system automatically calls the special nuclear power welder qualification rule engine module to intelligently analyze the sub-item decomposition, dispatching and early warning data. However, the system cannot infer the specific parameters in the operation process, resulting in some difficult projects that cannot meet the high quality requirements by relying only on existing data for welding. Summary of the invention

[0005] The purpose of the present invention is to propose a nuclear power welding specification information fusion management system based on artificial intelligence in view of the existing deficiencies.

[0006] The present invention adopts the following technical solution:

[0007] An artificial intelligence-based nuclear power welding specification information fusion management system, including a knowledge base module, a data fusion module, an AI reasoning module, a user management module and a mobile terminal module;

[0008] The knowledge base module is used to store and manage knowledge resources of nuclear power welding, the data fusion module is used to integrate different welding processes and output welding plans, the AI ​​reasoning module is used to reason and analyze the processes in the welding plan and generate process parameters, the user management module is used to manage the qualification information and operation records of welders, and the mobile terminal module is used to query and feedback on on-site information;

[0009] The knowledge base module includes a standard information storage unit, an inference rule management unit and a knowledge retrieval unit, wherein the standard information storage unit is used to store standard information of nuclear power welding, the inference rule management unit is used to manage basic inference rule information, and the knowledge retrieval unit is used to retrieve standard information and rule information;

[0010] The data fusion module includes a vector management unit, a screening and integration unit, and a solution output unit. The vector management unit is used to convert the welding standard into a feature vector. The screening and integration unit screens the vectors based on the requirements and performs integration processing. The solution output unit is used to integrate the solutions for evaluation and output the best solution.

[0011] The AI ​​reasoning module includes a process disassembly unit, a reasoning analysis unit and a feedback optimization unit. The process disassembly unit is used to disassemble the welding plan into multiple process tasks. The reasoning analysis unit is used to perform reasoning calculations on the parameters in each process task. The feedback optimization unit is used to collect actual operation results and optimize the reasoning analysis unit.

[0012] The user management module includes a welder file management unit, an intelligent screening unit and a permission review unit. The welder file management unit is used to manage welder information, the intelligent screening unit is used to screen out welders for performing welding tasks, and the permission review unit is used to review and verify the welder's permissions;

[0013] Further, the vector management unit includes a vector generation processor, a vector mapping register and a similarity calculation processor, wherein the vector generation processor is used to convert the welding standard into a feature vector, the vector mapping register is used to store the standard vector and the mapping relationship between the feature vector and the standard information, and the similarity calculation processor is used to calculate the similarity between the requirement vector and the feature vector;

[0014] Furthermore, the similarity calculation processor calculates the similarity Sr between the feature vector and the demand vector according to the following formula:

[0015]

[0016] Among them, a i represents the value of the i-th element in the feature vector, b irepresents the value of the i-th element in the demand vector, k i represents the weight value of the i-th element, and n is the number of elements in the feature vector and the demand vector;

[0017] Furthermore, the process disassembly unit includes a step disassembly processor, a parameter filling processor and a task output processor, wherein the step disassembly processor is used to disassemble the welding standard into a plurality of process tasks, the parameter filling processor is used to fill the parameters determined in the requirement information into the process task, and the task output processor is used to control the process of outputting the process task to the reasoning analysis unit;

[0018] Furthermore, the process of the task output processor outputting the process task includes the following steps:

[0019] S21, sequentially obtain a process task as a target task;

[0020] S22, determine whether there are unknown parameters in the target task, if yes, go to step S24, if not, go to step S23;

[0021] S23, putting the target task into the execution queue, and returning to step S21;

[0022] S24, obtaining all process tasks from the execution queue as reasoning conditions, and sending the reasoning conditions and the target tasks together to the reasoning analysis unit;

[0023] S25, waiting to receive feedback of the target task, and checking whether there are any unknown parameters in the target task, if yes, directly reporting an error, if no, entering step S23.

[0024] The beneficial effects achieved by the present invention are:

[0025] This system can effectively integrate multiple welding standards and process specifications, support automated decision-making in complex scenarios, simulate the reasoning process of welding experts, provide efficient and accurate process design solutions, and verify the welder qualifications and welding process at the operation site to ensure that the welding task can be completed with high quality.

[0026] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0028] Figure 2 This is a schematic diagram of the data fusion module of the present invention;

[0029] Figure 3 This is a schematic diagram of the AI ​​reasoning module of the present invention;

[0030] Figure 4 This is a schematic diagram of the user management module of the present invention;

[0031] Figure 5 This is a schematic diagram of the mobile terminal module of the present invention;

[0032] Figure 6 It is the experimental test effect diagram of the present invention. DETAILED DESCRIPTION

[0033] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0034] Embodiment 1.

[0035] This embodiment provides a nuclear power welding specification information fusion management system based on artificial intelligence, combined with Figure 1 , including knowledge base module, data fusion module, AI reasoning module, user management module and mobile terminal module.

[0036] The knowledge base module is used to store and manage knowledge resources for nuclear power welding, the data fusion module is used to integrate different welding processes and output welding plans, the AI ​​reasoning module is used to reason and analyze the processes in the welding plan and generate process parameters, the user management module is used to manage welder qualification information and operation records, and the mobile terminal module is used to query and feedback on on-site information.

[0037] The knowledge base module includes a standard information storage unit, an inference rule management unit and a knowledge retrieval unit. The standard information storage unit is used to store standard information of nuclear power welding, the inference rule management unit is used to manage basic inference rule information, and the knowledge retrieval unit is used to retrieve standard information and rule information.

[0038] The data fusion module includes a vector management unit, a screening and integration unit and a solution output unit. The vector management unit is used to convert the welding standard into a feature vector. The screening and integration unit screens the vectors based on the requirements and performs integration processing. The solution output unit is used to integrate the solutions for evaluation and output the best solution.

[0039] The AI ​​reasoning module includes a process disassembly unit, a reasoning analysis unit and a feedback optimization unit. The process disassembly unit is used to disassemble the welding plan into multiple process tasks, the reasoning analysis unit is used to perform reasoning calculations on the parameters in each process task, and the feedback optimization unit is used to collect actual operation results and optimize the reasoning analysis unit.

[0040] The user management module includes a welder file management unit, an intelligent screening unit and an authority review unit. The welder file management unit is used to manage welder information, the intelligent screening unit is used to screen out welders for performing welding tasks, and the authority review unit is used to review and verify the welder's authority.

[0041] The vector management unit includes a vector generation processor, a vector mapping register and a similarity calculation processor. The vector generation processor is used to convert the welding standard into a feature vector. The vector mapping register is used to store the standard vector and the mapping relationship between the feature vector and the standard information. The similarity calculation processor is used to calculate the similarity between the requirement vector and the feature vector.

[0042] The similarity calculation processor calculates the similarity Sr between the feature vector and the requirement vector according to the following formula:

[0043]

[0044] Among them, a i represents the value of the i-th element in the feature vector, b i represents the value of the i-th element in the demand vector, k i represents the weight value of the i-th element, and n is the number of elements in the feature vector and the demand vector.

[0045] The process disassembly unit includes a step disassembly processor, a parameter filling processor and a task output processor. The step disassembly processor is used to disassemble the welding standard into multiple process tasks, the parameter filling processor is used to fill the parameters determined in the requirement information into the process task, and the task output processor is used to control the process of outputting the process task to the reasoning analysis unit.

[0046] The process of the task output processor outputting the process task includes the following steps:

[0047] S21. Obtain a process task in sequence as a target task.

[0048] S22. Determine whether there are unknown parameters in the target task. If so, proceed to step S24; if not, proceed to step S23.

[0049] S23, putting the target task into the execution queue and returning to step S21.

[0050] S24. Obtain all process tasks from the execution queue as reasoning conditions, and send the reasoning conditions and the target tasks together to the reasoning analysis unit.

[0051] S25, waiting to receive feedback of the target task, and checking whether there are any unknown parameters in the target task, if yes, directly reporting an error, if no, entering step S23.

[0052] Embodiment 2.

[0053] This embodiment includes all the contents of the first embodiment, and provides an artificial intelligence-based nuclear power welding specification information fusion management system, including a knowledge base module, a data fusion module, an AI reasoning module, a user management module and a mobile terminal module.

[0054] The knowledge base module is used to store and manage knowledge resources for nuclear power welding, the data fusion module is used to integrate different welding processes and output welding plans, the AI ​​reasoning module is used to reason and analyze the processes in the welding plan and generate process parameters, the user management module is used to manage welder qualification information and operation records, and the mobile terminal module is used to query and feedback on on-site information.

[0055] The knowledge base module includes a standard information storage unit, an inference rule management unit and a knowledge retrieval unit. The standard information storage unit is used to store standard information of nuclear power welding, the inference rule management unit is used to manage basic inference rule information, and the knowledge retrieval unit is used to retrieve standard information and rule information.

[0056] Combination Figure 2 The data fusion module includes a vector management unit, a screening and integration unit and a solution output unit. The vector management unit is used to convert the welding standard into a feature vector. The screening and integration unit screens the vector based on the requirements and performs integration processing. The solution output unit is used to integrate the solutions for evaluation and output the best solution.

[0057] Combination Figure 3The AI ​​reasoning module includes a process disassembly unit, a reasoning analysis unit and a feedback optimization unit. The process disassembly unit is used to disassemble the welding plan into multiple process tasks, the reasoning analysis unit is used to perform reasoning calculations on the parameters in each process task, and the feedback optimization unit is used to collect actual operation results and optimize the reasoning analysis unit.

[0058] Combination Figure 4 The user management module includes a welder file management unit, an intelligent screening unit and an authority review unit. The welder file management unit is used to manage welder information, the intelligent screening unit is used to screen out welders for performing welding tasks, and the authority review unit is used to review and verify the welder's authority.

[0059] Combination Figure 5 The mobile terminal module includes an identity authentication unit, a process monitoring unit and a synchronous feedback unit. The identity authentication unit is used to verify the identity of the welder, the process monitoring unit is used to monitor and verify the real-time operation of the welder, and the synchronous feedback unit is used to synchronously feedback the field data to the user management module.

[0060] The standard information storage unit includes a welding standard information register and a process standard information register. The welding standard information register is used to store different versions of basic welding requirements, and the process standard information register is used to store detailed operating procedure cases.

[0061] The inference rule management unit includes a rule base, a rule editing processor and a rule triggering processor. The rule base is used to store rules and regulations, the rule editing processor is used to edit the rules and regulations in the rule base, and the rule triggering processor is used to call the rules and regulations for the module to perform decision reasoning.

[0062] The knowledge retrieval unit includes a keyword analysis processor, an information matching processor and a knowledge recommendation processor. The keyword analysis processor is used to analyze and process the keywords in the search content, the information matching processor is used to match the analysis results with the content in the standard information storage unit and the rule base, and the knowledge recommendation processor recommends and displays the matching results.

[0063] The vector management unit includes a vector generation processor, a vector mapping register and a similarity calculation processor. The vector generation processor is used to convert the welding standard into a feature vector. The vector mapping register is used to store the standard vector and the mapping relationship between the feature vector and the standard information. The similarity calculation processor is used to calculate the similarity between the requirement vector and the feature vector.

[0064] The screening and integration unit includes a demand analysis processor and an integration control processor. The demand analysis processor is used to receive demand information and parse to obtain a corresponding demand vector. The integration control processor screens feature vectors from the vector management unit based on the demand vector and combines them into a solution.

[0065] The process of building the integrated control processor solution includes the following steps:

[0066] S1. Obtain feature vectors from the vector management unit.

[0067] S2. Calling a similarity calculation processor to calculate the similarity between the feature vector and the requirement vector.

[0068] S3. Filter and retain feature vectors based on similarity values.

[0069] S4. Repeat steps S1 to S3 to complete the screening based on all demand vectors.

[0070] S5. Combine the feature vectors to form a solution.

[0071] The similarity calculation processor calculates the similarity Sr between the feature vector and the requirement vector according to the following formula:

[0072]

[0073] Among them, a i represents the value of the i-th element in the feature vector, b i represents the value of the i-th element in the demand vector, k i represents the weight value of the i-th element, and n is the number of elements in the feature vector and the demand vector.

[0074] The solution output unit includes a solution evaluation processor and a conversion output processor. The solution evaluation processor is used to evaluate the comprehensive adaptability of the solution, and the conversion output processor is used to convert the vector combination into a form and output it to the AI ​​reasoning module.

[0075] The solution evaluation processor calculates the adaptation index At according to the following formula:

[0076]

[0077] Among them, m is the number of demand vectors, δ(i, i+1) represents the connection constant between the welding standard corresponding to the i-th feature vector in the scheme and the welding standard corresponding to the i+1-th feature vector, and Sr(i) represents the similarity between the i-th feature vector in the scheme and the corresponding demand vector.

[0078] The connection constant is greater than or equal to 1, and the specific value of the connection index between the two welding standards is set by the technicians according to the actual situation.

[0079] The conversion output processor converts and outputs the solution with the smallest adaptation index.

[0080] The process disassembly unit includes a step disassembly processor, a parameter filling processor and a task output processor. The step disassembly processor is used to disassemble the welding standard into multiple process tasks, the parameter filling processor is used to fill the parameters determined in the requirement information into the process task, and the task output processor is used to control the process of outputting the process task to the reasoning analysis unit.

[0081] The process of the task output processor outputting the process task includes the following steps:

[0082] S21. Obtain a process task in sequence as a target task.

[0083] S22. Determine whether there are unknown parameters in the target task. If so, proceed to step S24; if not, proceed to step S23.

[0084] S23, putting the target task into the execution queue and returning to step S21.

[0085] S24. Obtain all process tasks from the execution queue as reasoning conditions, and send the reasoning conditions and the target tasks together to the reasoning analysis unit.

[0086] S25, waiting to receive feedback of the target task, and checking whether there are any unknown parameters in the target task, if yes, directly reporting an error, if no, entering step S23.

[0087] The reasoning analysis unit includes a rule acquisition processor, a rule application processor and a reasoning feedback processor. The rule acquisition processor is used to obtain rule regulations from the reasoning rule management unit. The rule application processor performs reasoning calculations on unknown parameters in the target task based on the rule regulations. The reasoning feedback processor is used to fill the reasoned parameters into the target task and then feed them back to the process disassembly unit.

[0088] The rule application processor calculates the probability of the target task according to the following formula:

[0089]

[0090] Among them, w i represents the i-th unknown parameter value in the target task, P(w i |w1,w2,···,w i-1 ) means that when there are unknown parameters w1, w2, ..., wi-1 Under the condition that the i-th unknown parameter value is w i The probability of n is the number of unknown parameters in the target task.

[0091] The rule application processor obtains the probabilities under different parameter conditions by changing the value of the unknown parameter and selects the condition with the highest probability to determine the value of the unknown parameter.

[0092] The welder file management unit includes a static information register and a dynamic information register. The static information register is used to store static information such as the welder's name, photo, contact information, qualification certificate, and qualification validity period. The dynamic information register is used to store dynamic information such as the welder's operation record, idle status, and current address.

[0093] The intelligent screening unit includes a static screening processor and a dynamic screening processor. The static screening processor performs preliminary screening on welders based on the matching relationship between welder qualifications and projects, and the dynamic screening processor screens out welder information for executing projects based on the current status of the welders.

[0094] The identity authentication unit includes an identity information register, an image capture processor and an image recognition processor. The identity information register is used to receive and save the screened welder information. The image capture processor is used to capture the welder's facial information. The image recognition processor is used to proofread and verify the captured image with the head portrait in the identity information.

[0095] The process monitoring unit includes a task information register, an operation monitoring processor and a process identification processor. The task information register is used to store the process task information of the project, the operation monitoring processor is used to shoot the operation process of the welder, and the process identification processor is used to perform action recognition on the operation process in the video and determine whether it meets the process task.

[0096] The i in the above text is an ordinal number used to indicate a sequence number and has no actual meaning.

[0097] Some of the code for this system is described below:

[0098]

[0099]

[0100]

[0101]

[0102] We are testing 10 projects, using this system and a common system to manage and process them respectively, recording the completion time of each project, and obtaining Figure 6 The effect comparison chart is shown.

[0103] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A nuclear power welding specification information fusion management system based on artificial intelligence, characterized in that: It includes knowledge base module, data fusion module, AI reasoning module, user management module and mobile terminal module; The knowledge base module is used to store and manage knowledge resources of nuclear power welding, the data fusion module is used to integrate different welding processes and output welding plans, the AI ​​reasoning module is used to reason and analyze the processes in the welding plan and generate process parameters, the user management module is used to manage the qualification information and operation records of welders, and the mobile terminal module is used to query and feedback on on-site information; The knowledge base module includes a standard information storage unit, an inference rule management unit and a knowledge retrieval unit, wherein the standard information storage unit is used to store standard information of nuclear power welding, the inference rule management unit is used to manage basic inference rule information, and the knowledge retrieval unit is used to retrieve standard information and rule information; The data fusion module includes a vector management unit, a screening and integration unit, and a solution output unit. The vector management unit is used to convert the welding standard into a feature vector. The screening and integration unit screens the vectors based on the requirements and performs integration processing. The solution output unit is used to integrate the solutions for evaluation and output the best solution. The AI ​​reasoning module includes a process disassembly unit, a reasoning analysis unit and a feedback optimization unit. The process disassembly unit is used to disassemble the welding plan into multiple process tasks. The reasoning analysis unit is used to perform reasoning calculations on the parameters in each process task. The feedback optimization unit is used to collect actual operation results and optimize the reasoning analysis unit. The user management module includes a welder file management unit, an intelligent screening unit and an authority review unit. The welder file management unit is used to manage welder information, the intelligent screening unit is used to screen out welders for performing welding tasks, and the authority review unit is used to review and verify the welder's authority.

2. The artificial intelligence-based nuclear power welding specification information fusion management system according to claim 1, characterized in that: The vector management unit includes a vector generation processor, a vector mapping register and a similarity calculation processor. The vector generation processor is used to convert the welding standard into a feature vector. The vector mapping register is used to store the standard vector and the mapping relationship between the feature vector and the standard information. The similarity calculation processor is used to calculate the similarity between the requirement vector and the feature vector.

3. The artificial intelligence-based nuclear power welding specification information fusion management system as claimed in claim 2, characterized in that: The similarity calculation processor calculates the similarity Sr between the feature vector and the requirement vector according to the following formula: Among them, a i represents the value of the i-th element in the feature vector, b i represents the value of the i-th element in the demand vector, k i represents the weight value of the i-th element, and n is the number of elements in the feature vector and the demand vector.

4. The artificial intelligence-based nuclear power welding specification information fusion management system as claimed in claim 3, characterized in that: The process disassembly unit includes a step disassembly processor, a parameter filling processor and a task output processor. The step disassembly processor is used to disassemble the welding standard into multiple process tasks, the parameter filling processor is used to fill the parameters determined in the requirement information into the process task, and the task output processor is used to control the process of outputting the process task to the reasoning analysis unit.

5. The artificial intelligence-based nuclear power welding specification information fusion management system as claimed in claim 4, characterized in that: The process of the task output processor outputting the process task includes the following steps: S21, sequentially obtain a process task as a target task; S22, determine whether there are unknown parameters in the target task, if yes, go to step S24, if not, go to step S23; S23, putting the target task into the execution queue, and returning to step S21; S24, obtaining all process tasks from the execution queue as reasoning conditions, and sending the reasoning conditions and the target tasks together to the reasoning analysis unit; S25, waiting to receive feedback of the target task, and checking whether there are any unknown parameters in the target task, if yes, directly reporting an error, if no, entering step S23.

Images