Dangerous chemical substance safety production administrative approval method based on knowledge graph and rule base

Abstract

The invention is suitable for the technical field of knowledge graph and administrative approval, and provides a dangerous chemical substance safety production administrative approval method based on a knowledge graph and a rule base, which comprises the following steps: collecting original data of dangerous chemical substance safety production administrative approval, and generating an original database of dangerous chemical substance safety production administrative approval, constructing a dangerous chemical safety production administrative approval knowledge graph; setting a rule base of hazardous chemical substance safety production administrative approval; according to the submitted approval application request, the knowledge graph is called according to the request content to search the corresponding approval data, the approval data and the set rule base are judged one by one, when all the approval data meet the requirements of the rule base, it is judged that the current approval is passed, and intelligent approval of safety production of the hazardous chemical substances is achieved. Based on the established knowledge graph and rule base, the hazardous chemical substance safety production administrative examination and approval method is provided, intelligent examination and approval of hazardous chemical substance safety production administrative examination and approval are achieved, the manual processing cost is reduced, and the examination and approval efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of safety management, and in particular to a method for administrative approval of hazardous chemicals production safety based on a knowledge graph and a rule base. Background Art

[0002] Hazardous chemicals refer to chemicals with flammable, explosive, toxic, corrosive and other characteristics. There are high safety risks in their production, storage, transportation and use. In order to ensure the safety of the public and the environment, countries have formulated relevant laws and regulations to regulate the production and use of hazardous chemicals, including the hazardous chemicals production safety system. Under this system, companies need to go through a series of approval procedures before they can legally produce or use hazardous chemicals.

[0003] However, since the administrative approval for hazardous chemicals production safety involves multiple links and departments, the process is cumbersome, the information is complex, and it requires a lot of knowledge of laws and regulations.

[0004] Traditional approval methods are manual approval methods, which often have the following problems: low approval efficiency, irregular operation, difficulty in ensuring compliance, and prone to human errors. At present, there is no computer-based intelligent approval method for hazardous chemicals in patent searches. Therefore, the intelligent administrative approval method for safe production of hazardous chemicals is of great significance in ensuring that the approval process complies with relevant laws and regulations, and improving approval compliance and user satisfaction. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide a method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base, aiming to solve the technical problems of high labor cost, complex laws and regulations, and low approval efficiency in administrative approval of hazardous chemicals production safety.

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

[0007] Step S1: collect the original data of administrative approval for production safety of hazardous chemicals, extract entities, relationships and attributes from the original data, generate the original database of administrative approval for production safety of hazardous chemicals, and construct a knowledge graph of administrative approval for production safety of hazardous chemicals;

[0008] Step S2: Setting a rule base for administrative approval of hazardous chemicals production safety;

[0009] Step S3: Based on the submitted approval application request, the knowledge graph is called according to the request content to search for the corresponding approval data, and the data is judged one by one with the set rule base. When all the requirements of the rule base are met, the current approval is determined to be passed, thus realizing the intelligent approval of hazardous chemicals safety production.

[0010] Furthermore, the specific process of step S1 is as follows:

[0011] S11. Collect the original data of administrative approval for production safety of hazardous chemicals;

[0012] S12, performing data format conversion processing on the collected semi-structured raw data, including converting the raw data in docx format into plain text data in txt format, and converting the raw data in pdf format into image data;

[0013] S13, preprocessing the image data;

[0014] S14, performing text recognition on the pre-processed image data to obtain plain text data in txt format;

[0015] S15. Extract user interface element information from all plain text data, extract entities, relationships and attributes, and build the original database of administrative approval for safe production of hazardous chemicals through MySQL database;

[0016] S16. Based on the original database, a knowledge graph of administrative approval for production safety of hazardous chemicals is constructed through the neo4j graph database.

[0017] Furthermore, the process of image data preprocessing in step S13 is as follows: grayscale the image data, then perform geometric changes on the grayscaled image, scale it to a suitable uniform size, and finally perform data enhancement on the uniform size image.

[0018] Furthermore, the specific process of step S15 is as follows:

[0019] S151, extracting entities and building an entity database;

[0020] Configure entity tags according to requirements, extract relevant entity information according to entity tags, and then build entity data tables to obtain entity databases. The attributes of entity data tables include unique indexes, entity names, and parent entity IDs. Entity hierarchical relationships are built through parent entity IDs. If there is no parent entity, the attributes are empty.

[0021] S152, extracting entity attributes and building an entity attribute database;

[0022] Configure attribute tags according to requirements, extract relevant entity attribute information according to attribute tags, and then construct a tag attribute data table to obtain an entity attribute database, wherein the attributes included in the tag attribute data table include unique index, attribute name, tag id, definition domain and data type, wherein the definition domain is used to mark the initial entity to which the attribute belongs; the data type is represented by a number to distinguish various data types;

[0023] S153, extracting relations and constructing a relational database;

[0024] Configure relationship tags according to requirements, extract relevant relationship information according to relationship tags, and then construct a relationship data table to obtain a relationship database, where the relationship data table includes attributes such as unique index, relationship name, source tag id, target tag id, and definition domain;

[0025] S154. Extract relationship attributes and build a relationship attribute database.

[0026] Configure relationship attribute labels according to requirements, extract relevant relationship attribute information according to relationship attribute labels, and then build a relationship attribute data table to obtain a relationship attribute database, where the relationship attribute data table includes attributes such as unique index, attribute name, relationship id, data type, and definition domain.

[0027] Furthermore, in step S2, the set rule base includes a first application / extension rule base and a change rule base;

[0028] The rule set of the first application / extension rule base is as follows:

[0029] Rule 1: Requirements for documents and applications for safety production licenses;

[0030] Rule 2: Requirements for enterprise personnel arrangements and safety production responsibility system documents;

[0031] Rule 3: Rules and regulations;

[0032] Rule 4: Requirements for work safety regulations and operating procedures for each type of work;

[0033] Rule 5: Requirements for enterprises to have full-time production safety management personnel;

[0034] Rule 6: Rules and requirements of the enterprise administrative management body;

[0035] Rule 7: Rules and requirements for the educational background and qualifications of each person in the enterprise;

[0036] Rule 8: Requirements for companies to submit withdrawal and use reports;

[0037] Rule 9: Rules and requirements for enterprise work-related injury insurance and the number of insured persons;

[0038] Rule 10: Rules and requirements for corporate emergency rescue;

[0039] Rule 11: Rules and requirements for the list of rescue equipment, equipment and facilities;

[0040] Rule 12: The regulations and requirements for the varieties registered by the enterprise, the scope of application for license and the validity period;

[0041] Rule 13: Regulatory requirements for safety assessment reports;

[0042] Rule 14: Requirements for expert acceptance opinions, confirmation materials, and enterprise self-acceptance report;

[0043] Rule 15: Rules requirements for filing proof;

[0044] Rule 16: Rules requirements for operators;

[0045] The rule set for changing the rule base is as follows:

[0046] Rule 1: Corresponding rules and requirements for changing the company name;

[0047] Rule 2: Corresponding rules and requirements for changing the registered address;

[0048] Rule 3: Change the corresponding rules and requirements for the principal person in charge.

[0049] Furthermore, the specific process of step S3 is as follows:

[0050] S31. When a submitted approval application request is received, the knowledge graph is called to search for corresponding approval data according to the request content;

[0051] S32. According to the type of the submitted approval application request, a corresponding rule base is used. For an initial application or an extension application, an initial application / extension rule base is selected. For a change application, a change rule base is selected.

[0052] S33, executing judgment logic one by one on the approval data according to the rule set of the corresponding rule library, and when all the requirements of the rule library are met, the current approval is determined to be passed, otherwise the approval is not passed;

[0053] S34. Output the approval results and complete the intelligent approval of hazardous chemicals production safety.

[0054] Furthermore, the specific process of step S33 is: determine whether the approval data is passed in turn according to the judgment logic of each rule. If passed, add "1" to the rule judgment set, otherwise, add "0" to the rule judgment set; loop judgment until all rules are judged; if all elements in the rule judgment set are "1", the approval is passed, otherwise, the approval is not passed.

[0055] The beneficial effects of the present invention are:

[0056] First, in order to solve the problem of lack of data support and decision-making basis in the administrative approval process of hazardous chemicals production safety, the present invention constructs a hazardous chemicals production safety administrative approval database to form a hazardous chemicals production safety administrative approval knowledge graph. Through this knowledge graph, various information on the production safety of hazardous chemicals production enterprises can be quickly and accurately retrieved, which improves information transparency and approval efficiency.

[0057] In addition, in view of the lack of unified standardized guidance in the approval process, which is prone to inconsistency, the present invention integrates relevant laws and regulations, establishes a hazardous chemicals production safety administrative approval rule library, achieves compliance in the approval work, improves the accuracy and consistency of the approval, and reduces errors and risks in the approval process;

[0058] Finally, in view of the complicated approval process, high manual processing costs, and a large amount of time and human resources consumed, the present invention reduces manual processing costs, improves approval efficiency, and saves time and resources through the entire process design of the administrative approval method for safe production of hazardous chemicals. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a method for administrative approval of production safety of hazardous chemicals based on a knowledge graph and a rule base provided by an embodiment of the present invention;

[0060] Figure 2-Figure 20 It is a flowchart of the judgment logic of each rule of the rule library provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0063] like Figure 1 As shown, the administrative approval method for safe production of hazardous chemicals based on knowledge graph and rule base provided in this embodiment includes the following processes:

[0064] Step S1: Collect the original data of administrative approval for production safety of hazardous chemicals, extract entities, relationships and attributes from the original data, generate the original database of administrative approval for production safety of hazardous chemicals, and construct a knowledge graph of administrative approval for production safety of hazardous chemicals.

[0065] Knowledge graph is a method of expressing knowledge through a graph structure, which represents and stores various objects in the real world and their interrelationships through nodes (entities) and edges (relationships). Knowledge graph is a structured semantic knowledge base used to describe concepts and their interrelationships in the physical world, to achieve smarter search engines and to provide richer information retrieval functions. In the field of administrative approval for safe production of hazardous chemicals, the method for constructing a knowledge graph for administrative approval for safe production of hazardous chemicals provided in this embodiment is as follows:

[0066] S11. Collect the original data of administrative approval for production safety of hazardous chemicals;

[0067] S12, performing data format conversion processing on the collected semi-structured raw data, including converting the raw data in docx format into plain text data in txt format, and converting the raw data in pdf format into image data;

[0068] S13, preprocessing the image data, the main process is: graying the image data, then performing geometric changes on the grayed image, scaling it to a suitable uniform size, and finally performing data enhancement on the uniform size image;

[0069] S14, performing text recognition on the pre-processed image data to obtain plain text data in txt format;

[0070] S15. Perform UIE (user interface element) information extraction on all plain text data, extract entities, relationships and attributes, and build the original database of administrative approval for safe production of hazardous chemicals through MySQL database;

[0071] S16. Based on the original database, a knowledge graph of administrative approval for production safety of hazardous chemicals is constructed through the neo4j graph database.

[0072] Specifically, the process of establishing the original database in step S15 is as follows:

[0073] S151, extracting entities and building an entity database;

[0074] Configure entity labels according to requirements. For example, if you want to extract information about companies, personnel, and organizations, set schema = ['Company', 'Personnel', 'Organization', ...], extract relevant entity information according to entity labels, and then construct an entity data table (label) to obtain an entity database. The entity data table includes attributes such as unique index (id), entity name (title), and parent entity id (parent_id). Since entities have inheritance relationships, the entity hierarchy is constructed in the database through the parent entity id (parent_id). If there is no parent entity, the attribute is empty.

[0075] An example of an entity data table is as follows:

[0076] id title parent_id 1692360816308436993 enterprise Null 1693473145917960193 personnel Null 1693788699199422465 mechanism 1692360816308436993 …… …… ……

[0077] S152, extracting entity attributes and building an entity attribute database;

[0078] Configure attribute labels according to requirements. For example, to extract relevant attributes of an enterprise, set schema = {'Enterprise': ['Name', 'Type', 'Registered Address', ...]}, extract relevant entity attribute information according to the attribute labels, and then construct a label attribute data table (label_property) to obtain an entity attribute database, where the label attribute data table includes attributes such as unique index (id), attribute name (title), label id (label_id), domain (domain), and data type (type). Since the child entity will inherit all attributes of the parent entity, the domain attribute is set to mark the initial entity to which the attribute belongs, that is, the domain is used to mark the initial entity to which the attribute belongs. Since there are multiple types of data in entity attributes, the data type (type) is represented by a number to distinguish between data types, such as 0 for string type, 1 for int data, 2 for double data, 3 for datetime data, etc.

[0079] Taking the above sample as an example, the label attribute data table is as follows:

[0080]

[0081] S153, extracting relations and constructing a relational database;

[0082] Configure relationship tags according to requirements. For example, if you want to extract the relationship between an enterprise and a person, set schema = {'unit': ['main person in charge', 'security director', 'security administrator', ...]}; if you want to extract the relationship between an enterprise and an organization, set schema = {'unit': ['establishment']}. Extract relevant relationship information based on relationship tags, and then construct a relationship data table (relation) to obtain a relationship database. The relationship data table includes the following attributes: unique index (id), relationship name (title), source label id (source_label_id), target label id (target_label_id), and domain (domain).

[0083] Taking the above sample as an example, the relational data representation is as follows:

[0084]

[0085] S154. Extract relationship attributes and build a relationship attribute database.

[0086] Configure relationship attribute labels according to needs. For example, to extract attributes such as the time when an enterprise established an institution, set schema = {'establishment': ['time']}, extract relevant relationship attribute information according to the relationship attribute labels, and then build a relationship attribute data table (relation_property) to obtain a relationship attribute database, where the relationship attribute data table includes attributes such as unique index (id), attribute name (title), relationship id (relation_id), data type (type), and definition domain (domain).

[0087] Taking the above sample as an example, the relation_property data table is as follows:

[0088]

[0089] According to the above process and examples, it is possible to construct a knowledge graph of administrative approval for production safety of hazardous chemicals based on the original data of administrative approval for production safety of hazardous chemicals.

[0090] Step S2: Setting a rule base for administrative approval of hazardous chemicals production safety;

[0091] The rule bases that are set up include the first application / extension rule base and the change rule base. The rule requirements of the two types of rule bases are different. Among them, the rule bases for first application and extension have 16 rules, and the change rule base has 3 rules. The rule requirements are as follows:

[0092] The rule set for the Initial Application / Extension Rule Base is as follows:

[0093] Rule 1: Requirements for documents and applications for safety production licenses;

[0094] Rule 2: Requirements for enterprise personnel arrangements and safety production responsibility system documents;

[0095] Rule 3: Rules and regulations;

[0096] Rule 4: Requirements for work safety regulations and operating procedures for each type of work;

[0097] Rule 5: Requirements for enterprises to have full-time production safety management personnel;

[0098] Rule 6: Rules and requirements of the enterprise administrative management body;

[0099] Rule 7: Rules and requirements for the educational background and qualifications of each person in the enterprise;

[0100] Rule 8: Requirements for companies to submit withdrawal and use reports;

[0101] Rule 9: Rules and requirements for enterprise work-related injury insurance and the number of insured persons;

[0102] Rule 10: Rules and requirements for corporate emergency rescue;

[0103] Rule 11: Rules and requirements for the list of rescue equipment, equipment and facilities;

[0104] Rule 12: The regulations and requirements for the varieties registered by the enterprise, the scope of application for license and the validity period;

[0105] Rule 13: Regulatory requirements for safety assessment reports;

[0106] Rule 14: Requirements for expert acceptance opinions, confirmation materials, and enterprise self-acceptance report;

[0107] Rule 15: Rules requirements for filing proof;

[0108] Rule 16: Rules requirements for operators;

[0109] The rule set for changing the rule base is as follows:

[0110] Rule 1: Corresponding rules and requirements for changing the company name;

[0111] Rule 2: Corresponding rules and requirements for changing the registered address;

[0112] Rule 3: Change the corresponding rules and requirements for the principal person in charge.

[0113] For the above-mentioned rule requirements, as a specific method, taking the first application and extension of the safety production license of a hazardous chemical production enterprise as an example, the rule set example is formulated as follows (the judgment process diagram corresponding to rules 1-16 is as follows: Figure 2-17 shown):

[0114] Rule 1: The name, principal and address of the applicant in the documents and application form of the safety production license shall be consistent with the business license;

[0115] Rule 2: The enterprise shall have the main person in charge, the person in charge of the sub-personnel, other persons in charge, the person in charge of the enterprise functional management organization, staff, workshop (branch) team leaders, post workers, and the safety production responsibility system documents of each functional department, and shall be stamped with the official seal;

[0116] Rule 3: The enterprise shall formulate rules and regulations such as safety inspection system, occupational hazard prevention system, safety education and training system, production safety accident management system, major hazard source monitoring and major hidden danger rectification system, equipment safety management system, production safety file management system, production safety reward and punishment system, and affix the official seal;

[0117] Rule 4: The enterprise shall formulate work safety regulations and operating procedures for each type of work and affix the official seal;

[0118] Rule 5: The number of full-time safety production managers in an enterprise should be no less than 1% of the total number of employees, and at least 2 people; full-time safety production managers should obtain the qualification certificate of hazardous chemical safety managers. For enterprises with more than 500 employees, a hazardous chemical registered safety engineer or registered assistant safety engineer should be assigned at a ratio of no less than 30% of the safety production managers; for enterprises with less than 500 employees, at least 1 hazardous chemical registered safety engineer or registered assistant safety engineer should be assigned;

[0119] Rule 6: The enterprise shall issue a document in the form of an independent formal document to establish a safety production management organization and staff it with full-time safety production management personnel, and it should be an administrative management organization such as the safety department (section, unit);

[0120] Rule 7: The knowledge and management capabilities of the main person in charge, the person in charge of safety, and the safety production management personnel of the enterprise shall meet the requirements of the regulations, obtain valid certificates, and have certain chemical professional knowledge or more than 3 years of experience in the chemical industry. At least one of them shall have a bachelor's degree or above in chemical engineering and obtain a qualification certificate for hazardous chemical safety management personnel;

[0121] Rule 8: Enterprises shall submit a report on the withdrawal and use of security fees, stating the amount of security fees withdrawn by the enterprise in the previous year as a percentage of operating income, and the details of the use, and withdraw according to the following rules: if the operating income of the previous year does not exceed RMB 10 million, withdraw 4.5%; if the operating income of the previous year exceeds RMB 10 million and reaches RMB 100 million, withdraw 2.25%; if the operating income of the previous year exceeds RMB 100 million and reaches RMB 1 billion, withdraw 0.55%; if the operating income of the previous year exceeds RMB 1 billion, withdraw 0.2%;

[0122] Rule 9: The enterprise shall have proof of payment of work-related injury insurance for the past three months (tax receipts or invoices), and submit proof of all-employee liability insurance, and the number of insured persons shall be consistent with the number of employees in the application form;

[0123] Rule 10: The enterprise shall have formal documents to clearly define the emergency rescue organization (organizational structure, personnel division of labor, job responsibilities); smaller enterprises shall provide a list of emergency rescue personnel, work division, and contact information;

[0124] Rule 11: The company shall provide a list of rescue equipment, facilities and supplies;

[0125] Rule 12: The varieties registered by the enterprise are consistent with the scope of the application license and are within the validity period;

[0126] Rule 13: The cover of the safety evaluation report must be stamped with the official seal, the evaluation agency’s qualifications must be valid and have the evaluation capability, the list of evaluators must be signed, the evaluation report conclusion must be stamped, the technical person in charge and the person in charge of process control must be signed, and the evaluation report must be stamped with a saddle stamp;

[0127] Rule 14: The enterprise shall have the expert acceptance opinions, the materials signed and confirmed by the expert group leader after rectification, and the enterprise's self-acceptance report;

[0128] Rule 15: The enterprise has the name of the unit and the issuing authority of the record certificate, and the record certificate is within the validity period;

[0129] Rule 16: Operators of hazardous chemical processes in an enterprise must undergo professional training, obtain a valid special operations operation certificate, and work with the certificate. If they are reviewed, they must provide review records.

[0130] Taking the change of the safety production license of a hazardous chemical production enterprise as an example, the rule set is formulated as follows (the judgment process diagram corresponding to rules 1-3 is as follows Figure 18-20 shown):

[0131] Rule 1: If the company name is changed, the changed business license and its corresponding documents should be submitted;

[0132] Rule 2: If an enterprise changes its registered address during the validity period of its safety production license, it shall submit an application for change to the implementing authority within 10 working days from the date of change of its industrial and commercial business license or affiliation, and submit a copy of the changed industrial and commercial business license and relevant supporting materials;

[0133] Rule 3: If an enterprise changes its principal person in charge during the validity period of its safety production license, it shall submit an application for change to the implementing authority within 10 working days from the date of change of its industrial and commercial business license or affiliation, and submit a copy of the changed industrial and commercial business license and a copy of the safety certificate issued by the principal person in charge after passing the assessment by the safety production supervision and management department.

[0134] Step S3: Based on the submitted approval application request, the knowledge graph is called according to the request content to search for the corresponding approval data, and the data is judged one by one with the set rule base. When all the requirements of the rule base are met, the current approval is determined to be passed, thus realizing the intelligent approval of hazardous chemicals safety production.

[0135] This step is a specific approval process. The first user submits an approval application on the system. According to the submitted approval application request, the knowledge graph is called to search for data, and then the rule base is judged. If all the rules are passed, the approval is considered passed. The specific process of this step is as follows:

[0136] S31. When a submitted approval application request is received, the knowledge graph is called to search for corresponding approval data according to the request content;

[0137] S32. According to the type of the submitted approval application request, a corresponding rule base is used. For an initial application or an extension application, an initial application / extension rule base is selected. For a change application, a change rule base is selected.

[0138] S33. Execute the judgment logic one by one on the approval data according to the rule set of the corresponding rule library. When all the requirements of the rule library are met, the current approval is determined to be passed, otherwise the approval is not passed. For example, the approval data is judged in turn according to the judgment logic of each rule. If it is passed, "1" is added to the rule judgment set, otherwise, "0" is added to the rule judgment set; the judgment is repeated until all rules are judged; if all elements in the rule judgment set are "1", the approval is passed, otherwise, the approval is not passed.

[0139] S34. Output the approval results and complete the intelligent approval of hazardous chemicals production safety.

[0140] In summary, the present invention obtains a knowledge graph of administrative approval for production safety of hazardous chemicals by constructing an administrative approval database for production safety of hazardous chemicals, and constructs a rule base for administrative approval for production safety of hazardous chemicals according to relevant laws and regulations. Based on the constructed knowledge graph and rule base, an administrative approval method for production safety of hazardous chemicals is proposed to realize intelligent approval of administrative approval for production safety of hazardous chemicals, reduce manual processing costs, and improve approval efficiency.

[0141] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base, characterized in that: The method comprises the following steps: Step S1: collect the original data of administrative approval for production safety of hazardous chemicals, extract entities, relationships and attributes from the original data, generate the original database of administrative approval for production safety of hazardous chemicals, and construct a knowledge graph of administrative approval for production safety of hazardous chemicals; Step S2: Setting a rule base for administrative approval of hazardous chemicals production safety; Step S3: Based on the submitted approval application request, the knowledge graph is called according to the request content to search for the corresponding approval data, and the data is judged one by one with the set rule base. When all the requirements of the rule base are met, the current approval is determined to be passed, thus realizing the intelligent approval of hazardous chemicals safety production.

2. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 1 is characterized in that: The specific process of step S1 is as follows: S11. Collect the original data of administrative approval for production safety of hazardous chemicals; S12, performing data format conversion processing on the collected semi-structured raw data, including converting the raw data in docx format into plain text data in txt format, and converting the raw data in pdf format into image data; S13, preprocessing the image data; S14, performing text recognition on the pre-processed image data to obtain plain text data in txt format; S15. Extract user interface element information from all plain text data, extract entities, relationships and attributes, and build the original database of administrative approval for safe production of hazardous chemicals through MySQL database; S16. Based on the original database, a knowledge graph of administrative approval for production safety of hazardous chemicals is constructed through the neo4j graph database.

3. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 2 is characterized in that: The process of image data preprocessing in step S13 is as follows: grayscale the image data, then perform geometric changes on the grayscaled image, scale it to a suitable uniform size, and finally perform data enhancement on the uniform size image.

4. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 3 is characterized in that: The specific process of step S15 is as follows: S151, extracting entities and building an entity database; Configure entity tags according to requirements, extract relevant entity information according to entity tags, and then build entity data tables to obtain entity databases. The attributes of entity data tables include unique indexes, entity names, and parent entity IDs. Entity hierarchical relationships are built through parent entity IDs. If there is no parent entity, the attributes are empty. S152, extracting entity attributes and building an entity attribute database; Configure attribute tags according to requirements, extract relevant entity attribute information according to attribute tags, and then construct a tag attribute data table to obtain an entity attribute database, wherein the attributes included in the tag attribute data table include unique index, attribute name, tag id, definition domain and data type, wherein the definition domain is used to mark the initial entity to which the attribute belongs; the data type is represented by a number to distinguish various data types; S153, extracting relations and constructing a relational database; Configure relationship tags according to requirements, extract relevant relationship information according to relationship tags, and then construct a relationship data table to obtain a relationship database, where the relationship data table includes attributes such as unique index, relationship name, source tag id, target tag id, and definition domain; S154. Extract relationship attributes and build a relationship attribute database. Configure relationship attribute labels according to requirements, extract relevant relationship attribute information according to relationship attribute labels, and then build a relationship attribute data table to obtain a relationship attribute database, where the relationship attribute data table includes attributes such as unique index, attribute name, relationship id, data type, and definition domain.

5. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 4 is characterized in that: In step S2, the set rule base includes an initial application / extension rule base and a change rule base; The rule set of the first application / extension rule base is as follows: Rule 1: Requirements for documents and applications for safety production licenses; Rule 2: Requirements for enterprise personnel arrangements and safety production responsibility system documents; Rule 3: Rules and regulations; Rule 4: Requirements for work safety regulations and operating procedures for each type of work; Rule 5: Requirements for enterprises to have full-time production safety management personnel; Rule 6: Rules and requirements of the enterprise administrative management body; Rule 7: Rules and requirements for the educational background and qualifications of each person in the enterprise; Rule 8: Requirements for companies to submit withdrawal and use reports; Rule 9: Rules and requirements for enterprise work-related injury insurance and the number of insured persons; Rule 10: Rules and requirements for corporate emergency rescue; Rule 11: Rules and requirements for the list of rescue equipment, equipment and facilities; Rule 12: The regulations and requirements for the varieties registered by the enterprise, the scope of application for license and the validity period; Rule 13: Regulatory requirements for safety assessment reports; Rule 14: Requirements for expert acceptance opinions, confirmation materials, and enterprise self-acceptance report; Rule 15: Rules requirements for filing proof; Rule 16: Rules requirements for operators; The rule set for changing the rule base is as follows: Rule 1: Corresponding rules and requirements for changing the company name; Rule 2: Corresponding rules and requirements for changing the registered address; Rule 3: Change the corresponding rules and requirements for the principal person in charge.

6. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 5, characterized in that: The specific process of step S3 is as follows: S31. When a submitted approval application request is received, the knowledge graph is called to search for corresponding approval data according to the request content; S32. According to the type of the submitted approval application request, a corresponding rule base is used. For an initial application or an extension application, an initial application / extension rule base is selected. For a change application, a change rule base is selected. S33, executing judgment logic one by one on the approval data according to the rule set of the corresponding rule library, and when all the requirements of the rule library are met, the current approval is determined to be passed, otherwise the approval is not passed; S34. Output the approval results and complete the intelligent approval of hazardous chemicals production safety.

7. The method for administrative approval of hazardous chemicals production safety based on knowledge graph and rule base as claimed in claim 6 is characterized in that: The specific process of step S33 is: determine whether the approval data is passed in turn according to the judgment logic of each rule. If passed, add "1" to the rule judgment set, otherwise, add "0" to the rule judgment set; loop judgment until all rules are judged; if all elements in the rule judgment set are "1", the approval is passed, otherwise, the approval is not passed.

Images