A nuclear power plant isolation safety measure information automatic identification method based on semantic analysis
By using a semantic analysis-based method to automatically identify text corpora of nuclear power plant isolation safety measures, the problem of the difficulty in automatically identifying nuclear power plant isolation information has been solved, improving the identification efficiency of isolation information and the level of intelligent management, thus ensuring the safety and reliability of nuclear power units.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA NUCLEAR POWER OPERATION TECH CORP
- Filing Date
- 2022-12-19
- Publication Date
- 2026-05-15
AI Technical Summary
Unstructured or semi-structured text data required for isolation and safety measures in nuclear power plant work orders are difficult to automatically identify, resulting in low efficiency in identifying isolation information and an inability to effectively extract useful structured information, which affects the process of refining and intelligentizing nuclear power plant isolation management.
Using a semantic analysis-based approach, through Chinese word segmentation, part-of-speech tagging, noise reduction, disambiguation, semantic matching, and semantic dependency analysis, we can automatically identify isolation information in the text corpus of nuclear power plant isolation safety requirements, including whether isolation is required, isolation boundary equipment, and the status of the isolation target.
It has improved the efficiency of automatic identification of isolation information in nuclear power plants, reduced the risks of conflict and uncertainty in safety measures, enhanced the level of intelligence in production planning and management, and ensured the safety, availability and reliability of nuclear power units.
Smart Images
Figure CN116108135B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information recognition technology, and in particular to an automatic identification method for isolation safety measures in nuclear power plants based on semantic analysis. Background Technology
[0002] While work order management and isolation management in nuclear power plants have largely achieved informatization and process-orientation, the rapid development of digital and intelligent transformation in nuclear power and the strong drive for refined management have placed higher demands on work order management and isolation management. Currently, domestic and international research on nuclear power plant isolation mainly focuses on optimizing isolation measures and resolving conflicts based on input isolation information. The identification of isolation information in the description of isolation safety measures for nuclear power plant work orders still relies primarily on manual labor.
[0003] The isolation and safety measure requirements in nuclear power plant work orders consist of unstructured or semi-structured text data. These requirements involve systems, equipment, equipment status, and equipment relationships, and the text quality varies greatly. Accurately identifying valid information requires significant personnel experience and is inefficient. Currently, there is a lack of research on machine-automated recognition of isolation and safety measure requirements in nuclear power plant work orders. This makes it impossible to automatically extract useful structured isolation information from a large volume of nuclear power plant work orders, and the deeper value of understanding these requirements remains to be explored. Summary of the Invention
[0004] The purpose of this invention is to provide an automatic identification method for nuclear power plant isolation safety measures based on semantic analysis. This method enables the automatic identification of nuclear power plant isolation safety measures information such as "whether isolation is required", "isolation boundary equipment", and "isolation target status corresponding to isolation boundary equipment" in the description of isolation safety measures requirements in nuclear power plant work orders. This helps relevant professionals to understand the isolation status of the entire unit's systems and equipment within a specific time window.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] An automatic identification method for isolation safety measures in nuclear power plants based on semantic analysis includes the following steps:
[0007] Step 1: Categorize the textual corpus of nuclear power plant isolation and safety measures requirements;
[0008] Step 2: Preprocess the text corpus of nuclear power plant isolation and safety measures requirements;
[0009] Step 3: Based on semantic matching, automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated;
[0010] Step 4: Automatically identify the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus based on semantic dependency analysis;
[0011] Step 5: Based on semantic dependency analysis, automatically identify the isolation target status of the nuclear power plant work order task isolation boundary equipment corresponding to the nuclear power plant isolation safety measure requirement text corpus.
[0012] Step 6: Based on the automatic identification results of isolation safety measures in the nuclear power plant work order, achieve a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window.
[0013] Furthermore, in step 1, the nuclear power plant isolation and safety measure requirements text corpus consists of the content of the "isolation and safety measure requirements" field in the nuclear power plant work order task; the nuclear power plant isolation and safety measure requirements text corpus is sorted according to the description category.
[0014] Further, step 2 involves preprocessing the text corpus of nuclear power plant isolation and safety measures requirements, including the following steps:
[0015] Step 2.1: Chinese word segmentation of the text corpus on nuclear power plant isolation and safety measures requirements;
[0016] Step 2.2: Part-of-speech tagging of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements;
[0017] Step 2.3: Noise reduction processing of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements;
[0018] Step 2.4: Disambiguation processing of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements.
[0019] Further, step 2.1, the Chinese word segmentation of the nuclear power plant isolation safety measures requirements text corpus, includes the following steps:
[0020] Based on a general lexicon, new words are identified in a large corpus of nuclear power plant isolation and safety measures requirements texts using machine learning and manual annotation, in order to supplement, improve and form a professional lexicon of nuclear power plant isolation and safety measures.
[0021] Based on a specialized lexicon of nuclear power plant isolation and safety measures, Chinese word segmentation of text corpora containing nuclear power plant isolation and safety measures requirements is achieved through a conditional random field model combined with the Viterbi algorithm.
[0022] Further, step 2.2, part-of-speech tagging of the word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements, includes the following steps:
[0023] Based on the Chinese word segmentation results, the part-of-speech tags of the segmentation results are automatically labeled using a hidden Markov model. The part-of-speech tags include nouns, verbs, adjectives, adverbs, conjunctions, locative words, numerals, personal names, and noun-verbs.
[0024] Further, step 2.3, the noise reduction processing of the word segmentation results of the nuclear power plant isolation safety measures requirements text corpus, includes the following steps:
[0025] Based on the Chinese word segmentation results, noise is removed and reduced; the noise includes continuous punctuation marks and full-width and half-width characters.
[0026] Further, step 2.4, the disambiguation processing of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements, includes the following steps:
[0027] Based on the Chinese word segmentation results, the demonstrative pronouns are subjected to disambiguation processing.
[0028] Further, step 3, based on semantic matching, automatically identifies whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated, including the following steps:
[0029] Step 3.1: Manually annotate the text corpus of nuclear power plant isolation safety measures requirements with the question mark "whether isolation is required";
[0030] Step 3.2: Perform semantic matching between the text corpus of nuclear power plant isolation safety measures requirements to be processed and the text corpus of labeled isolation safety measures requirements, and automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements to be processed needs to be isolated;
[0031] Step 3.3: If the identification result is that the nuclear power plant isolation safety measures requirement text corpus corresponding to the nuclear power plant work order task needs to be isolated, then proceed to step 4.
[0032] If the identification result is that the nuclear power plant work order task corresponding to the text corpus of isolation safety measures requirements for the nuclear power plant to be processed does not need to be isolated, then skip steps 4 and 5 and proceed directly to step 6.
[0033] Further, step 3.2 involves semantically matching the text corpus of isolation and safety measures requirements for nuclear power plants to be processed with the text corpus of labeled isolation and safety measures requirements, automatically identifying whether the work order task corresponding to the text corpus of isolation and safety measures requirements needs to be isolated, including the following steps:
[0034] Based on the characteristics of the different types of isolation and safety measure requirement text corpus in step 1 and the preprocessing results of the nuclear power plant isolation and safety measure requirement text corpus in step 2, the text corpus of nuclear power plant isolation and safety measure requirement to be processed is semantically similar to the text corpus of isolation and safety measure requirement already labeled, and the results of whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation and safety measure requirement to be processed needs to be isolated are obtained.
[0035] Further, step 4, based on semantic dependency analysis, automatically identifies the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirement text corpus, including the following steps:
[0036] Step 4.1: Based on the preprocessing results of the nuclear power plant isolation safety measure requirement text corpus from Step 2, perform semantic dependency graph representation on the unprocessed nuclear power plant isolation safety measure requirement text corpus corresponding to the work order tasks for nuclear power plants requiring isolation identified in Step 3.
[0037] Step 4.2: Locate the given predicate verb pointing to "isolation boundary equipment" in the text corpus of the nuclear power plant isolation safety measures requirements corresponding to the work order task for isolating nuclear power plants identified in Step 3;
[0038] Step 4.3: Based on the semantic roles of the given predicate verbs in the semantic dependency graph, automatically identify the isolation boundary equipment involved in the work order task that requires the isolation of the nuclear power plant identified in Step 3.
[0039] Furthermore, in step 4.1, the isolation safety measures for nuclear power plants require that the semantic dependency relations of the text corpus include verb-object relations, object-complement relations, attributive-head relations, and prepositional-object relations.
[0040] Furthermore, in step 4.2, the given predicate verbs in the nuclear power plant isolation safety measures requirement text corpus that point to "isolation boundary equipment" include, but are not limited to: "interlocking", "isolation", and "shutting down".
[0041] Furthermore, in step 4.3, the semantic roles on which the given predicate verbs “interlocking,” “isolation,” and “closing” depend are generally the dependent object nouns “device name” and “device code”, forming a verb-object relationship, thereby automatically identifying the isolation boundary equipment involved in the work order task associated with the given predicate verb.
[0042] Further, step 5, based on semantic dependency analysis, automatically identifies the isolation target state corresponding to the isolation boundary equipment of the nuclear power plant work order task, which is based on the nuclear power plant isolation safety measure requirement text corpus. This includes the following steps:
[0043] Step 5.1: Based on the isolation boundary equipment involved in the work order task of the nuclear power plant that needs to be isolated identified in Step 4, locate the given object noun pointing to "isolation target state" in the text corpus of the nuclear power plant isolation safety measures requirements to be processed corresponding to the work order task of the nuclear power plant that needs to be isolated identified in Step 3.
[0044] Step 5.2: Based on the semantic state of the given object noun in the semantic dependency graph in step 4, automatically identify the isolation target state corresponding to the isolation boundary equipment of the work order task that needs to be isolated from the nuclear power plant identified in step 3.
[0045] Furthermore, in step 5.1, the given object noun pointing to the “isolation target state” in the nuclear power plant isolation safety measures requirement text corpus is the name of the isolation boundary equipment identified in step 4.
[0046] Furthermore, in step 5.12, the semantic state on which the given object noun depends is a dependent object complement, which includes, but is not limited to, "shutdown" and "operation status", that is, the isolation target state corresponding to the isolation boundary device.
[0047] Furthermore, step 6, based on the automatic identification results of isolation safety measures in the nuclear power plant's work orders, achieves a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window, including the following steps:
[0048] By integrating the automatic identification results of whether the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 3 need to be isolated, the automatic identification results of the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 4, and the automatic identification results of the isolation target status corresponding to the isolation boundary equipment of the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 5, a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window can be achieved.
[0049] The present invention also provides an automatic identification device for nuclear power plant isolation safety measures based on semantic analysis, the device comprising:
[0050] processor;
[0051] Memory used to store processor-executable instructions;
[0052] The processor is configured to execute the above method.
[0053] The present invention also provides a non-volatile computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the above-described method.
[0054] Beneficial technical effects of the present invention:
[0055] The present invention provides an automatic identification method for nuclear power plant isolation safety measures based on semantic analysis. This method achieves automatic identification of nuclear power plant isolation safety measures based on semantic analysis, which greatly improves the identification efficiency of relevant professionals in identifying isolation information of nuclear power plant work orders and tasks. It lays the foundation for improving the intelligence level of the isolation preparation process and production planning management process, reduces the uncertainty risks such as safety measure contradictions, temporary SPV restrictions and I0 conflicts caused by insufficient identification of isolation information, and reduces the unavailability rate of SSCs caused by shutdown maintenance, thereby maximizing the safety, availability and reliability of nuclear power unit systems and equipment. Attached Figure Description
[0056] Figure 1 This is a flowchart of the automatic identification method for nuclear power plant isolation safety measures based on semantic analysis according to the present invention.
[0057] Figure 2 Example diagram of semantic dependency in text corpus for isolation and safety measures requirements of nuclear power plants. Detailed Implementation
[0058] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0059] See Figure 1 This invention provides an automatic identification method for nuclear power plant isolation safety measures based on semantic analysis, comprising the following steps:
[0060] Step 1: Categorize the textual corpus of nuclear power plant isolation and safety measures requirements;
[0061] Step 2: Preprocess the text corpus of nuclear power plant isolation and safety measures requirements;
[0062] Step 3: Based on semantic matching, automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated;
[0063] Step 4: Automatically identify the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus based on semantic dependency analysis;
[0064] Step 5: Based on semantic dependency analysis, automatically identify the isolation target status of the nuclear power plant work order task isolation boundary equipment corresponding to the nuclear power plant isolation safety measure requirement text corpus.
[0065] Step 6: Based on the automatic identification results of isolation safety measures in the nuclear power plant work order, achieve a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window.
[0066] Furthermore, in step 1, the nuclear power plant isolation and safety measure requirement text corpus is the content of the "Isolation and Safety Measure Requirement" field in the nuclear power plant work order task; the nuclear power plant isolation and safety measure requirement text corpus is sorted according to the description category, which includes the "Interlock + Alarm" type, the "Isolate XX Equipment" type, the "Close XX Equipment Isolation Valve" type, and the "Isolate XX Equipment Pipeline" type.
[0067] Further, step 2 involves preprocessing the text corpus of nuclear power plant isolation and safety measures requirements, including the following steps:
[0068] Step 2.1: Perform Chinese word segmentation on the text corpus of the nuclear power plant isolation safety measures requirements.
[0069] Step 2.2: Perform词性标注 on the results of Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements.
[0070] Step 2.3: Perform noise reduction processing on the results of Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements.
[0071] Step 2.4: Perform disambiguation processing on the results of Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements.
[0072] Furthermore, for Step 2.1, the Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements includes the following steps:
[0073] Based on the general word library, use machine learning and manual annotation methods to identify new words in a large amount of text corpus of the nuclear power plant isolation safety measures requirements, so as to supplement, improve and form a professional word library for the nuclear power plant isolation safety measures.
[0074] Based on the professional word library for the nuclear power plant isolation safety measures, use the conditional random field model combined with the Viterbi algorithm to achieve Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements.
[0075] Furthermore, for Step 2.2, the词性标注 of the results of word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements includes the following steps:
[0076] Based on the results of Chinese word segmentation, use the hidden Markov model to automatically annotate the词性 of the word segmentation results, and the词性 includes nouns, verbs, adjectives, adverbs, conjunctions, locative words, numerals, personal names and nominal verbs.
[0077] Furthermore, for Step 2.3, the noise reduction processing of the results of word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements includes the following steps:
[0078] Based on the results of Chinese word segmentation, perform cleaning and noise reduction processing on the existing noise; the noise includes continuous punctuation marks and full-width and half-width symbols.
[0079] Furthermore, for Step 2.4, the disambiguation processing of the results of Chinese word segmentation of the text corpus of the nuclear power plant isolation safety measures requirements includes the following steps:
[0080] Based on the results of Chinese word segmentation, perform指代消歧 processing on the demonstrative pronouns therein. For example, judge the specific information of the system and equipment indicated by the demonstrative pronoun "该" in the descriptions of "该段管段", "该阀门" and "该设备".
[0081] It should be noted that the specific terms "词性标注" and "指代消歧" in the original text may need to be accurately translated according to the specific context and professional knowledge. Here, rough translations are provided for reference.Further, step 3, based on semantic matching, automatically identifies whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated, including the following steps:
[0082] Step 3.1: Manually annotate the text corpus of nuclear power plant isolation safety measures requirements to indicate whether isolation is required;
[0083] Step 3.2: Perform semantic matching between the text corpus of nuclear power plant isolation safety measures requirements to be processed and the text corpus of labeled isolation safety measures requirements, and automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements to be processed needs to be isolated;
[0084] Step 3.3: If the identification result is that the nuclear power plant work order task corresponding to the text corpus of isolation safety measures requirements for the nuclear power plant to be processed needs to be isolated, then proceed to step 4.
[0085] If the identification result is that the nuclear power plant work order task corresponding to the text corpus of isolation safety measures requirements for the nuclear power plant to be processed does not need to be isolated, then skip steps 4 and 5 and proceed directly to step 6.
[0086] Further, step 3.2 involves semantically matching the text corpus of isolation and safety measures requirements for nuclear power plants to be processed with the text corpus of labeled isolation and safety measures requirements, automatically identifying whether the work order task corresponding to the text corpus of isolation and safety measures requirements needs to be isolated, including the following steps:
[0087] Based on the characteristics of the different types of isolation and safety measure requirement text corpus in step 1 and the preprocessing results of the nuclear power plant isolation and safety measure requirement text corpus in step 2, the text corpus of nuclear power plant isolation and safety measure requirement to be processed is semantically similar to the text corpus of isolation and safety measure requirement already labeled, and the results of whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation and safety measure requirement to be processed needs to be isolated are obtained.
[0088] Further, step 4, based on semantic dependency analysis, automatically identifies the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirement text corpus, including the following steps:
[0089] Step 4.1, see Figure 2 Based on the preprocessing results of the nuclear power plant isolation safety measure requirement text corpus in step 2, semantic dependency graph representation is performed on the nuclear power plant isolation safety measure requirement text corpus corresponding to the work order task of the nuclear power plant to be isolated identified in step 3.
[0090] Step 4.2: Locate the given predicate verb pointing to "isolation boundary equipment" in the text corpus of the nuclear power plant isolation safety measures requirements corresponding to the work order task for isolating nuclear power plants identified in Step 3;
[0091] Step 4.3: Based on the semantic roles of the given predicate verbs in the semantic dependency graph, automatically identify the isolation boundary equipment involved in the work order task that requires the isolation of the nuclear power plant identified in Step 3.
[0092] Furthermore, in step 4.1, the text corpus of nuclear power plant isolation and safety measures generally lacks a subject, and the semantic dependency relations of the text corpus of nuclear power plant isolation and safety measures include verb-object relations, object-complement relations, attributive-head relations, and prepositional-object relations.
[0093] Furthermore, in step 4.2, the given predicate verbs in the nuclear power plant isolation safety measures requirement text corpus that point to "isolation boundary equipment" include, but are not limited to: "interlocking", "isolation", and "shutting down".
[0094] Furthermore, in step 4.3, the semantic roles on which the given predicate verbs “interlocking,” “isolation,” and “closing” depend are generally the dependent object nouns “device name” and “device code”, forming a verb-object relationship, thereby automatically identifying the isolation boundary equipment involved in the work order task associated with the given predicate verb.
[0095] Further, step 5, based on semantic dependency analysis, automatically identifies the isolation target state corresponding to the isolation boundary equipment of the nuclear power plant work order task, which is based on the nuclear power plant isolation safety measure requirement text corpus. This includes the following steps:
[0096] Step 5.1: Based on the isolation boundary equipment involved in the work order task of the nuclear power plant that needs to be isolated identified in Step 4, locate the given object noun pointing to "isolation target state" in the text corpus of the nuclear power plant isolation safety measures requirements to be processed corresponding to the work order task of the nuclear power plant that needs to be isolated identified in Step 3.
[0097] Step 5.2: Based on the semantic state of the given object noun in the semantic dependency graph in step 4, automatically identify the isolation target state corresponding to the isolation boundary equipment of the work order task that needs to be isolated from the nuclear power plant identified in step 3.
[0098] Furthermore, in step 5.1, the given object noun pointing to the “isolation target state” in the nuclear power plant isolation safety measures requirement text corpus is the name of the isolation boundary equipment identified in step 4.
[0099] Furthermore, in step 5.12, the semantic state on which the given object noun depends is generally a dependent object complement, which includes, but is not limited to, "shutdown" and "operation status", that is, the isolation target state corresponding to the isolation boundary device.
[0100] Furthermore, step 6, based on the automatic identification results of isolation safety measures in the nuclear power plant's work orders, achieves a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window, including the following steps:
[0101] By integrating the automatic identification results of whether the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 3 need to be isolated, the automatic identification results of the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 4, and the automatic identification results of the isolation target status corresponding to the isolation boundary equipment of the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 5, a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window can be achieved.
[0102] The present invention also provides an automatic identification device for nuclear power plant isolation safety measures based on semantic analysis, the device comprising:
[0103] processor;
[0104] Memory used to store processor-executable instructions;
[0105] The processor is configured to execute the above method.
[0106] The present invention also provides a non-volatile computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the above-described method.
[0107] Semantic analysis technology comprehensively mines the semantic relationships between different words such as subject, predicate, and object in natural language sentences, transforming natural language text corpora into formal language that reflects the meaning of sentences, enabling machines to understand their useful information.
[0108] By researching key technologies for automatic identification of isolation safety measures in nuclear power plants based on semantic analysis, this study addresses limitations in work order management and isolation management, such as the tendency for isolation managers to overlook isolation instructions and the inconvenience of planning engineers having to share or stagger time windows when assigning work order tasks. This not only improves the efficiency of work order and isolation management in power plants and reduces the workload of personnel, truly achieving "cost reduction and efficiency improvement," but also provides more solutions for the nuclear power industry to move towards intelligence and smart technology, ensuring the safe, reliable, and economical operation of nuclear power units and safeguarding the active, safe, and orderly development of nuclear power.
[0109] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.
Claims
1. A method for automatic identification of isolation safety measures information in nuclear power plants based on semantic analysis, characterized in that, Includes the following steps: Step 1: Categorize the textual corpus of nuclear power plant isolation and safety measures requirements; Step 2: Preprocessing the text corpus of nuclear power plant isolation and safety measures requirements; the preprocessing includes: based on a general lexicon, using machine learning and manual annotation to identify new words in a large amount of nuclear power plant isolation and safety measures requirement text corpus, forming a professional lexicon of nuclear power plant isolation and safety measures; based on the professional lexicon of nuclear power plant isolation and safety measures, using a conditional random field model combined with the Viterbi algorithm to achieve Chinese word segmentation of the nuclear power plant isolation and safety measures requirement text corpus; Step 3: Based on semantic matching, automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated; Step 4: Automatically identify the isolation boundary equipment involved in the nuclear power plant work order task corresponding to the nuclear power plant isolation safety measure requirement text corpus based on semantic dependency analysis; the semantic dependency analysis includes: locating the given predicate verb in the text corpus that points to "isolation boundary equipment", the given predicate verb includes "interlocking", "isolation" and "closing"; automatically identifying the isolation boundary equipment according to the dependent semantic roles of the given predicate verb in the semantic dependency graph, wherein the dependent semantic roles of the given predicate verb are the dependent object nouns "equipment name" and "equipment code", forming a verb-object relationship; Step 5: Automatically identify the isolation target state corresponding to the isolation boundary equipment of the nuclear power plant work order task based on the text corpus of nuclear power plant isolation safety measures requirements; the automatic identification of the isolation target state includes: identifying the semantic states that are dependent on the given object noun pointing to the "isolation target state" in the semantic dependency graph, wherein the semantic states are the dependent object complements "shutdown" and "operation status"; Step 6: Based on the automatic identification results of isolation safety measures in the nuclear power plant work order, achieve a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window.
2. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 1, characterized in that, In step 1, the nuclear power plant isolation and safety measure requirements text corpus consists of the content of the "Isolation and Safety Measure Requirements" field in the nuclear power plant work order task; the nuclear power plant isolation and safety measure requirements text corpus is sorted according to the description category.
3. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 1, characterized in that, Step 2 involves preprocessing the text corpus of nuclear power plant isolation and safety measures requirements, including the following steps: Step 2.1: Chinese word segmentation of the text corpus on nuclear power plant isolation and safety measures requirements; Step 2.2: Part-of-speech tagging of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements; Step 2.3: Noise reduction processing of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements; Step 2.4: Disambiguation processing of the Chinese word segmentation results of the text corpus on the requirements for isolation and safety measures of nuclear power plants.
4. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 3, characterized in that, Step 2.2 involves part-of-speech tagging of the word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements, including the following steps: Based on the Chinese word segmentation results, the part-of-speech tags of the segmentation results are automatically labeled using a hidden Markov model. The part-of-speech tags include nouns, verbs, adjectives, adverbs, conjunctions, locative words, numerals, personal names, and noun-verbs.
5. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 3, characterized in that, Step 2.3, noise reduction processing of the word segmentation results of the text corpus on nuclear power plant isolation safety measures requirements, includes the following steps: Based on the Chinese word segmentation results, noise is removed and reduced; the noise includes continuous punctuation marks and full-width and half-width characters.
6. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 3, characterized in that, Step 2.4, disambiguation processing of the Chinese word segmentation results of the text corpus on nuclear power plant isolation and safety measures requirements, includes the following steps: Based on the Chinese word segmentation results, the demonstrative pronouns are subjected to disambiguation processing.
7. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 1, characterized in that, Step 3: Based on semantic matching, automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements needs to be isolated, including the following steps: Step 3.1: Manually annotate the text corpus of nuclear power plant isolation safety measures requirements with the question mark "whether isolation is required"; Step 3.2: Perform semantic matching between the text corpus of nuclear power plant isolation safety measures requirements to be processed and the text corpus of labeled isolation safety measures requirements, and automatically identify whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation safety measures requirements to be processed needs to be isolated; Step 3.3: If the identification result is that the nuclear power plant isolation safety measures requirement text corpus corresponding to the nuclear power plant work order task needs to be isolated, then proceed to step 4. If the identification result is that the nuclear power plant work order task corresponding to the text corpus of isolation safety measures requirements for the nuclear power plant to be processed does not need to be isolated, then skip steps 4 and 5 and proceed directly to step 6.
8. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 7, characterized in that, Step 3.2 involves semantically matching the text corpus of nuclear power plant isolation and safety measures requirements to be processed with the text corpus of labeled isolation and safety measures requirements to automatically identify whether the work order task corresponding to the text corpus of isolation and safety measures requirements needs to be isolated. This includes the following steps: Based on the characteristics of the different types of isolation and safety measure requirement text corpus in step 1 and the preprocessing results of the nuclear power plant isolation and safety measure requirement text corpus in step 2, the text corpus of nuclear power plant isolation and safety measure requirement to be processed is semantically similar to the text corpus of isolation and safety measure requirement already labeled, and the results of whether the nuclear power plant work order task corresponding to the text corpus of nuclear power plant isolation and safety measure requirement to be processed needs to be isolated are obtained.
9. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 1, characterized in that, Step 4: Based on semantic dependency analysis, automatically identify the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirement text corpus, including the following steps: Step 4.1: Based on the preprocessing results of the nuclear power plant isolation safety measure requirement text corpus from Step 2, perform semantic dependency graph representation on the unprocessed nuclear power plant isolation safety measure requirement text corpus corresponding to the work order tasks for nuclear power plants requiring isolation identified in Step 3. Step 4.2: Locate the given predicate verb pointing to "isolation boundary equipment" in the text corpus of the nuclear power plant isolation safety measure requirements corresponding to the work order task for isolating nuclear power plants identified in Step 3; Step 4.3: Based on the semantic roles of the given predicate verbs in the semantic dependency graph, automatically identify the isolation boundary equipment involved in the work order task that requires the isolation of the nuclear power plant identified in Step 3.
10. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 9, characterized in that, In step 4.1, the isolation and safety measures for nuclear power plants require that the semantic dependency relations of the text corpus include verb-object relations, object-complement relations, attributive-head relations, and prepositional-object relations.
11. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 10, characterized in that, In step 5, the given object noun in the nuclear power plant isolation safety measures requirement text corpus that points to the "isolation target state" is the name of the isolation boundary equipment identified in step 4.
12. The method for automatic identification of nuclear power plant isolation safety measures information based on semantic analysis according to claim 1, characterized in that, Step 6: Based on the automatic identification results of isolation safety measures in the nuclear power plant work orders, achieve a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window, including the following steps: By integrating the automatic identification results of whether the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 3 need to be isolated, the automatic identification results of the isolation boundary equipment involved in the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 4, and the automatic identification results of the isolation target status corresponding to the isolation boundary equipment of the nuclear power plant work order tasks corresponding to the nuclear power plant isolation safety measure requirements text corpus in step 5, a comprehensive understanding of the system and equipment isolation status of the entire unit within a specific time window can be achieved.
13. An automatic identification device for isolation safety measures in nuclear power plants based on semantic analysis, characterized in that, The device includes: processor; Memory used to store processor-executable instructions; The processor is configured to perform the method according to any one of claims 1-12.
14. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1-12.