Maintenance scheme automatic generation method and system

Abstract

The invention discloses a method for automatically generating a maintenance scheme. The method comprises the following steps: establishing a maintenance scheme standard library; obtaining maintenance content of to-be-maintained equipment, inputting the maintenance content into the trained maintenance content semantic scene model, and outputting maintenance classification and maintenance part types; based on the maintenance classification and the maintenance part type, matching with the maintenance scheme standard library to generate a maintenance scheme; the time cost can be effectively reduced, and the overhaul scheme compiling efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and maintenance, and in particular to a method and system for automatically generating a maintenance plan. Background Art

[0002] With the continuous updating and development of power grid equipment, the structures of power grids at all levels are becoming more and more complex, and targeted safety measures are required. The maintenance plan is an important part of ensuring the safe and stable operation of equipment. Its core lies in systematically inheriting surveys, plans and other relevant data, and conducting in-depth integration, analysis and processing, so as to provide comprehensive and accurate guidance for the maintenance process.

[0003] The preparation of traditional maintenance plans is highly dependent on manual work. Based on on-site surveys and personal experience, it is necessary to comprehensively consider the process arrangements, potential risks, required equipment and materials, and coordination of on-site personnel for each operating surface. This process is extremely cumbersome and consumes a lot of time and human resources. Especially when facing a substation with many devices and a wide distribution, it is difficult to ensure the comprehensiveness of risk analysis based on personal experience alone, and it is easy to miss key risk points and achieve comprehensive coverage of risk prediction for all operating surfaces. Therefore, a method that can significantly improve the efficiency and accuracy of maintenance plan preparation is needed. Summary of the invention

[0004] The embodiment of the present invention provides a method for automatically generating a maintenance plan, which can improve the efficiency and accuracy of compiling a maintenance plan.

[0005] In a first aspect, an embodiment of the present invention provides a method for automatically generating a maintenance plan, comprising:

[0006] Establish a standard library of maintenance plans;

[0007] Obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type;

[0008] Based on the maintenance classification and the maintenance component type, matching is performed with the maintenance plan standard library to generate a maintenance plan.

[0009] Furthermore, the maintenance plan standard library includes an operation project process and risk identification standard library and a safety measures standard library.

[0010] Furthermore, the maintenance classification includes overall replacement and comprehensive shutdown, partial replacement and component repair, routine assessment and maintenance, conventional maintenance without power outage, and live maintenance without power outage.

[0011] Furthermore, the construction process of the maintenance content semantic scene model includes:

[0012] Obtaining historical maintenance content and preprocessing the historical maintenance content;

[0013] Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis;

[0014] Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

[0015] Furthermore, the maintenance classification and the maintenance component type are matched with the maintenance solution standard library to generate a maintenance solution, including:

[0016] According to the maintenance classification and the maintenance component type, based on the preset matching rules, matching is performed with the maintenance solution standard library to obtain a matching result; wherein the matching result includes at least one of the operation process, process risk level, process control measures, risk prevention measures and required tools;

[0017] A maintenance plan for the equipment to be repaired is generated according to the matching result.

[0018] In a second aspect, an embodiment of the present invention provides a system for automatically generating a maintenance plan, including:

[0019] Standard library establishment module, used to establish a standard library for maintenance plans;

[0020] A semantic analysis module is used to obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type;

[0021] A maintenance plan generation module is used to generate a maintenance plan by matching the maintenance classification and the maintenance component type with the maintenance plan standard library.

[0022] Furthermore, the maintenance plan standard library includes an operation project process and risk identification standard library and a safety measures standard library.

[0023] Furthermore, the maintenance classification includes overall replacement and comprehensive shutdown, partial replacement and component repair, routine assessment and maintenance, conventional maintenance without power outage, and live maintenance without power outage.

[0024] Furthermore, the construction process of the maintenance content semantic scene model includes:

[0025] Obtaining historical maintenance content and preprocessing the historical maintenance content;

[0026] Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis;

[0027] Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

[0028] Furthermore, the maintenance plan generating module includes:

[0029] A maintenance content matching unit, configured to match the maintenance classification and the maintenance component type with the maintenance solution standard library based on a preset matching rule to obtain a matching result; wherein the matching result includes at least one of an operation process, a process risk level, a process control measure, a risk prevention measure, and a required tool;

[0030] A maintenance plan generating unit is used to generate a maintenance plan for the equipment to be repaired according to the matching result.

[0031] Compared with the prior art, the method for automatically generating maintenance plans provided by the embodiment of the present invention has the following beneficial effects: by establishing a standard library of maintenance plans; obtaining the maintenance content of the equipment to be maintained, inputting the maintenance content into the trained semantic scene model of the maintenance content, and outputting the maintenance classification and maintenance component type; matching with the standard library of maintenance plans based on the maintenance classification and the maintenance component type, and generating a maintenance plan; significantly reducing the reporting time on a large amount of survey work content, thereby effectively reducing the time cost, and when new information of the equipment to be maintained is transmitted to the system, it can be quickly and accurately matched to the most similar situation, whether it is dealing with maintenance tasks of common equipment or responding to emergency repairs under special circumstances, it can provide effective solutions more quickly, which not only greatly improves the efficiency of maintenance plan compilation, but also enhances the accuracy of the plan, laying a solid foundation for the smooth progress of equipment maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical features of the embodiments of the present invention, the drawings required for use in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 It is a flowchart of a method for automatically generating a maintenance plan according to an embodiment of the present invention;

[0034] Figure 2 It is a structural schematic diagram of a maintenance plan automatic generation system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] In a first aspect, an embodiment of the present invention provides a method for automatically generating a maintenance plan, see Figure 1 , which is a flow chart of an embodiment of a method for automatically generating a maintenance plan provided by the present invention.

[0037] like Figure 1 As shown, the method comprises the following steps:

[0038] S1: Establish a standard library of maintenance plans;

[0039] S2: Obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type;

[0040] S3: Based on the maintenance classification and the maintenance component type, matching is performed with the maintenance plan standard library to generate a maintenance plan.

[0041] In the specific implementation, we first establish a standard library of maintenance plans, obtain the maintenance content of the equipment to be maintained through weekly plans or on-site survey orders, and the semantic scene model of the maintenance content. We use machine learning algorithms to train the semantic scene model to improve its recognition accuracy, and then input the acquired maintenance content into the trained semantic scene model of maintenance content. The model parses the maintenance content, identifies the semantic scene of maintenance, and outputs the maintenance classification and maintenance component type. Based on the maintenance classification and maintenance component type, we search for matches in the standard library of maintenance plans, and generate customized maintenance plans for the equipment to be maintained based on the matching results.

[0042] In summary, the present invention establishes a standard library of maintenance plans; obtains the maintenance content of the equipment to be repaired, inputs the maintenance content into the trained semantic scene model of the maintenance content, and outputs the maintenance classification and the type of maintenance components; based on the maintenance classification and the type of maintenance components, matches with the standard library of maintenance plans to generate a maintenance plan; it can significantly reduce the reporting time on a large amount of survey work content, thereby effectively reducing the time cost. When new information of the equipment to be repaired is transmitted to the system, it can quickly and accurately match the most similar situation. Whether it is dealing with maintenance tasks of common equipment or responding to emergency repairs under special circumstances, it can provide effective solutions more quickly, which not only greatly improves the efficiency of maintenance plan compilation, but also enhances the accuracy of the plan, laying a solid foundation for the smooth progress of equipment maintenance work.

[0043] In an optional implementation, the maintenance plan standard library includes an operation item process and risk identification standard library and a safety measures standard library.

[0044] Specifically, establishing a standard library for maintenance plans is a systematic and structured process, which aims to improve the efficiency, safety and standardization of maintenance operations. The standard library for maintenance plans mainly includes a standard library for operation project procedures and risk identification and a standard library for safety measures. Among them, the standard library for operation project procedures and risk identification mainly classifies maintenance operation projects according to factors such as equipment type, maintenance content, and technical difficulty. For each operation project, the process is refined according to the operation flow, and the specific operation content, sequence and technical requirements of each step are clarified. For each process, risk identification is carried out to identify possible safety hazards, sources of danger and potential accident types, and the identified risks are evaluated. According to the severity and probability of risk, the priority and control measures are determined. Based on the above contents, a standard library for operation project procedures and risk identification is established. The establishment of a standard library for safety measures requires the formulation of general safety measures applicable to all maintenance operations, special safety measures for specific operation projects or processes, emergency response plans, and safety training and education plans.

[0045] Establishing a standard library of maintenance plans can ensure that when preparing maintenance plans, the maintenance plans are less dependent on the proficiency of personnel's professional safety skills, as well as uncertain factors such as work attitude, to avoid problems such as imperfect and incorrect safety measures. Using a standard library for structured information management is more conducive to the standardization of maintenance plans, so that maintenance plans can be better adapted to different types of preparation, thereby improving the scalability of the system and the safety and accuracy of on-site operation risk management.

[0046] In an optional embodiment, the maintenance classification includes overall replacement and complete shutdown, partial replacement and component repair, routine assessment and maintenance, conventional maintenance without power outage, and live maintenance without power outage.

[0047] Specifically, the maintenance categories include A, B, C, D and E, which are respectively overall replacement and full shutdown, partial replacement and component repair, routine assessment and maintenance, routine maintenance without power outage and live maintenance. When the maintenance content contains natural languages ​​such as "overall replacement", "overall replacement of equipment", "full shutdown", "disassembly and replacement of the entire equipment or system" or "restart and debug and test the equipment", the maintenance category output by the maintenance content semantic scene model should be A. When the maintenance content contains natural languages ​​such as "partial replacement" or "component replacement", the maintenance category output by the maintenance content semantic scene model should be The maintenance classification should be Class B. When the maintenance content includes natural language such as "routine maintenance" or "conventional maintenance", it is usually carried out during equipment operation, so a comprehensive assessment of the equipment status is required, including the appearance of natural language such as "cleaning", "lubrication" or "tightening". The maintenance classification output by the maintenance content semantic scenario model should be Class C. When the maintenance content is a maintenance plan without power outages and no "live" related information appears, the maintenance classification output by the maintenance content semantic scenario model should be Class D. When the maintenance content is a maintenance plan without power outages and "live" related information appears, the maintenance classification output by the maintenance content semantic scenario model should be Class E.

[0048] Furthermore, the maintenance content is input into the maintenance content semantic scene model, and the type of maintenance components that need to be maintained must be output. For example, when the natural language "143 line PT" appears in the maintenance content, the maintenance component type output by the maintenance content semantic scene model should be voltage transformer; when the natural language "142-7PD knife switch mechanism component replacement" appears in the maintenance content, the maintenance component type output by the maintenance content semantic scene model should be valve body.

[0049] It should be noted that the maintenance classification and maintenance component types can be adjusted and set by the staff according to the actual work scenario and specific needs, and no specific restrictions are made here.

[0050] In an optional implementation, the process of constructing the maintenance content semantic scene model includes:

[0051] Obtaining historical maintenance content and preprocessing the historical maintenance content;

[0052] Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis;

[0053] Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

[0054] Specifically, when constructing a semantic scene model for maintenance content, we first obtain historical maintenance content and perform preprocessing operations on it, including data cleaning and annotation. Natural language processing technology is used to perform word vectorization and semantic relevance analysis on the preprocessed historical maintenance content. Based on the results of semantic analysis, a semantic scene model for maintenance content that can reflect the actual situation of maintenance activities is constructed. This can effectively transform maintenance content data into a model with rich semantic information, providing strong support for intelligent maintenance management.

[0055] In an optional implementation, the matching of the maintenance classification and the maintenance component type with the maintenance solution standard library to generate a maintenance solution includes:

[0056] According to the maintenance classification and the maintenance component type, based on the preset matching rules, matching is performed with the maintenance solution standard library to obtain a matching result; wherein the matching result includes at least one of the operation process, process risk level, process control measures, risk prevention measures and required tools;

[0057] A maintenance plan for the equipment to be repaired is generated according to the matching result.

[0058] Specifically, according to the maintenance classification and maintenance component type output by the maintenance content semantic scenario model, the maintenance plan standard library can be searched and compared according to the preset matching rules to find the matching result that best suits the current situation. This matching result covers multiple key information, including but not limited to the operation process (i.e. the steps that need to be performed during the maintenance process), process risk level (assessment of the degree of risk that each step may bring), process control measures (specific measures to ensure process quality and safety), risk prevention measures (strategies to prevent potential risks) and required tools (equipment or instruments necessary to complete the maintenance task). According to the matching results, a customized maintenance plan for the equipment to be maintained can be generated. The maintenance plan will list the maintenance process of each step in detail, clarify the risk level of each step and the corresponding control and prevention measures, as well as the tools required to be prepared, so as to ensure the smooth progress of the maintenance work and minimize potential risks.

[0059] In a second aspect, an embodiment of the present invention provides a system for automatically generating a maintenance plan, see Figure 2 As shown, it is a structural diagram of a maintenance plan automatic generation system provided by an embodiment of the present invention.

[0060] like Figure 2 As shown, the system comprises:

[0061] A standard library building module 21 is used to build a maintenance plan standard library;

[0062] The semantic analysis module 22 is used to obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type;

[0063] The maintenance plan generating module 23 is used to generate a maintenance plan by matching the maintenance classification and the maintenance component type with the maintenance plan standard library.

[0064] Furthermore, the maintenance plan standard library includes an operation project process and risk identification standard library and a safety measures standard library.

[0065] Furthermore, the maintenance classification includes overall replacement and comprehensive shutdown, partial replacement and component repair, routine assessment and maintenance, conventional maintenance without power outage, and live maintenance without power outage.

[0066] Furthermore, the construction process of the maintenance content semantic scene model includes:

[0067] Obtaining historical maintenance content and preprocessing the historical maintenance content;

[0068] Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis;

[0069] Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

[0070] Furthermore, the maintenance plan generating module includes:

[0071] A maintenance content matching unit, configured to match the maintenance classification and the maintenance component type with the maintenance solution standard library based on a preset matching rule to obtain a matching result; wherein the matching result includes at least one of an operation process, a process risk level, a process control measure, a risk prevention measure, and a required tool;

[0072] A maintenance plan generating unit is used to generate a maintenance plan for the equipment to be repaired according to the matching result.

[0073] The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. It should be pointed out that for those skilled in the art, several equivalent obvious variations and / or equivalent substitutions can be made without departing from the technical principles of the present invention. These obvious variations and / or equivalent substitutions should also be regarded as the protection scope of the present invention.

Claims

1. A method for automatically generating a maintenance plan, characterized in that: include: Establish a standard library of maintenance plans; Obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type; Based on the maintenance classification and the maintenance component type, matching is performed with the maintenance plan standard library to generate a maintenance plan.

2. The method for automatically generating a maintenance plan according to claim 1, characterized in that: The maintenance plan standard library includes an operation project process and risk identification standard library and a safety measures standard library.

3. The method for automatically generating a maintenance plan according to claim 1, characterized in that: The maintenance categories include overall replacement and complete shutdown, partial replacement and component repair, routine assessment and maintenance, regular maintenance without power outage, and live maintenance without power outage.

4. The method for automatically generating a maintenance plan according to claim 1, characterized in that: The construction process of the maintenance content semantic scene model includes: Obtaining historical maintenance content and preprocessing the historical maintenance content; Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis; Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

5. The method for automatically generating a maintenance plan according to claim 1, characterized in that: The method of matching the maintenance classification and the maintenance component type with the maintenance solution standard library to generate a maintenance solution includes: According to the maintenance classification and the maintenance component type, based on the preset matching rules, matching is performed with the maintenance solution standard library to obtain a matching result; wherein the matching result includes at least one of the operation process, process risk level, process control measures, risk prevention measures and required tools; A maintenance plan for the equipment to be repaired is generated according to the matching result.

6. A maintenance plan automatic generation system, characterized in that: include: Standard library establishment module, used to establish a standard library for maintenance plans; A semantic analysis module is used to obtain the maintenance content of the equipment to be repaired, input the maintenance content into the trained maintenance content semantic scene model, and output the maintenance classification and maintenance component type; A maintenance plan generation module is used to generate a maintenance plan by matching the maintenance classification and the maintenance component type with the maintenance plan standard library.

7. The automatic generation system of maintenance plan according to claim 6, characterized in that: The maintenance plan standard library includes an operation project process and risk identification standard library and a safety measures standard library.

8. The automatic generation system of maintenance plan according to claim 6, characterized in that: The maintenance categories include overall replacement and complete shutdown, partial replacement and component repair, routine assessment and maintenance, regular maintenance without power outage, and live maintenance without power outage.

9. The automatic generation system of maintenance plan according to claim 6, characterized in that: The construction process of the maintenance content semantic scene model includes: Obtaining historical maintenance content and preprocessing the historical maintenance content; Using natural language processing technology, the pre-processed historical maintenance content is subjected to word vectorization and semantic relevance analysis; Based on the results of semantic relevance analysis, a semantic scenario model of maintenance content is constructed.

10. The automatic generation system of maintenance plan according to claim 6, characterized in that: The maintenance plan generation module includes: A maintenance content matching unit, configured to match the maintenance classification and the maintenance component type with the maintenance solution standard library based on a preset matching rule to obtain a matching result; wherein the matching result includes at least one of an operation process, a process risk level, a process control measure, a risk prevention measure, and a required tool; A maintenance plan generating unit is used to generate a maintenance plan for the equipment to be repaired according to the matching result.

Images