A safety hazard data processing method and system based on an HSE intelligent agent

Abstract

A kind of safety hazard data processing method based on HSE agent, using external non-invasive architecture to operate HSE system, the method is executed by HSE agent, including: by bottom layer debugging protocol is attached to the browser instance that has logged in, takes over the interactive control right with the HSE system;Cross-system acquisition hazard operation data using three-layer degradation strategy;The hazard operation data obtained across systems is converged to central database, cross-table association chain is established by key field, cross-checking is carried out on the hazard operation data of multiple systems based on the cross-table association chain using rule engine, and verification state field is obtained;According to the verification state field, judge whether to build the fault work order of the hazard operation data in the browser instance;If the verification state field is established, then according to the preset attribution relationship, execute three-level fuzzy matching strategy and distribute the fault work order to the responsible unit;Realize low cost, short online cycle, the effect of seamless connection with original system.

Description

Technical Field

[0001] This invention belongs to the field of information management technology for safe production, and specifically relates to a method, system, storage medium, electronic device and program product for processing safety hazard data based on HSE intelligent agents. Background Technology

[0002] Existing safety hazard management systems face a common technical dilemma in achieving automated management of multi-source data fusion. If an "intrusive" modification is adopted, deep customization of the underlying code and databases of each independent HSE system (non-routine operations, major hazard sources, personnel location, etc.) is required. This is technically challenging, time-consuming, and compromises the stability and independence of the original systems. If relying on standard API interfaces, data acquisition failures often occur due to system version iterations, interface rate limiting, and SPA application caching mechanisms. This makes it impossible to achieve robust collection and consistency verification of multi-source heterogeneous data without zero intrusion into the existing systems. This technical dilemma is a deep-seated technical obstacle hindering the improvement of safety management automation levels.

[0003] Specifically, this manifests in the following ways: First, existing HSE hazard management processes heavily rely on manual operations, including manual registration of hazard information, manual assignment of rectification tasks, manual tracking of rectification progress, and manual review of rectification results. Manual operations are inefficient, prone to omissions, and difficult to guarantee standardized execution throughout the entire process. Second, existing safety hazard management systems all require enterprises to independently develop or purchase new software platforms. These solutions necessitate system redeployment, data migration, and personnel training, resulting in long implementation cycles, high costs, and difficulty in seamlessly integrating with existing HSE systems. Third, enterprises often have deployed multiple HSE-related systems, such as non-routine operation systems, major hazard source systems, and personnel positioning systems. Data is scattered across different platforms, making unified verification of hazard data across systems impossible and resulting in high costs. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a security vulnerability data processing method based on an HSE intelligent agent. This method employs an external, non-intrusive architecture to operate the HSE system. The method is executed by the HSE intelligent agent and includes the following steps:

[0005] S1. Attach to the logged-in browser instance through the underlying debugging protocol and take over the interaction control with the original HSE system;

[0006] S2. Use a three-tier degradation strategy to collect hidden danger operation data across systems;

[0007] S3. Aggregate the hidden danger operation data obtained from the cross-system to the central database, establish cross-table association chains through key fields, and use the rule engine to perform cross-validation on the hidden danger operation data of multiple systems based on the cross-table association chains to obtain the validation status field.

[0008] S4. Determine whether to construct a fault work order for the hidden danger operation data in the browser instance based on the verification status field; if the verification status field is valid, then execute a three-level fuzzy matching strategy according to the preset attribution relationship to dispatch the fault work order to the responsible unit.

[0009] S5. Repeat steps S2-S4 according to the preset cycle.

[0010] Preferably, the three-layer degradation strategy includes: prioritizing the interception of backend API responses and classifying and identifying them according to data structure characteristics; if the SPA cache hit causes the inability to obtain the real request, then forcibly triggering a page reload instruction to obtain it.

[0011] If the retrieval fails, return to the page document and retrieve the hazard operation data through DOM model parsing;

[0012] When online acquisition of the hazard operation data fails, the browser instance is controlled to force the specified download path to export offline files and parse them, thereby collecting hazard operation data across systems.

[0013] Preferably, the returned page document obtains the hidden danger operation data through DOM model parsing, including: locating and extracting page elements through a combination of preset multi-component library adaptation rules and multi-selectors, wherein the multi-component library adaptation rules are used to ensure compatibility with different front-end UI component libraries.

[0014] Preferably, the three-level fuzzy matching strategy includes: performing precise matching based on the department field in the hidden danger operation data and a preset responsibility unit mapping table; when precise matching fails, performing fuzzy matching based on the hidden danger category field in the hidden danger operation data and a preset business affiliation rule; when fuzzy matching fails, dispatching the fault work order corresponding to the hidden danger operation data to the superior affiliation unit according to a preset default rule.

[0015] Preferably, the step of establishing a cross-table association chain through key fields and using a rule engine to perform cross-validation on the hazard operation data of multiple systems based on the cross-table association chain includes: extracting the work ticket number, source identifier, and company name from the non-routine operation system, the major hazard source system, and the personnel positioning system as the key fields; establishing the cross-table association chain by collecting the key fields and cleaning and associating the fields across systems; and performing cross-system cross-validation based on the cross-table association chain through the rule engine. The cross-validation includes comparing the safety measure status in the work ticket with the actual alarm handling status. When it is detected that a safety measure is selected as not involved or the alarm has expired and not been handled, it is determined that there is a safety hazard.

[0016] Preferably, after the fault work order is dispatched to the responsible unit, the rectification status of the hidden dangers corresponding to the fault work order is reviewed periodically according to a preset review cycle; when the review result is not passed, the reviewed object is taken as the new work order responsible object, a new fault work order is generated and dispatched; when the hidden danger operation data processing is detected to be completed, the fault work order is automatically closed.

[0017] In a second aspect, the present invention provides a security vulnerability data processing system based on HSE intelligent agents, which operates the HSE system using an external non-intrusive architecture. The system includes: a protocol attachment module, which is used to attach to a logged-in browser instance through a low-level debugging protocol and take over the interaction control with the HSE system.

[0018] The degradation acquisition module is used to collect hazard operation data across systems using a three-layer degradation strategy;

[0019] The data aggregation and verification module is used to aggregate the hidden danger operation data obtained from cross systems to the central database, establish cross-table association chains through key fields, and use the rule engine to perform cross-verification on the associated data of multiple systems to obtain the verification status field.

[0020] The task construction and dispatch module is used to determine whether to construct the fault work order corresponding to the hidden danger data in the browser instance based on the verification status field, and dispatch the fault work order corresponding to the hidden danger operation data to the responsible unit according to the preset attribution relationship and the three-level fuzzy matching strategy.

[0021] The periodic scheduling module is used to trigger the repeated execution of the degradation acquisition module, data aggregation and verification module, and task construction and dispatch module according to a preset period.

[0022] A third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0023] A fourth aspect of the present invention provides an electronic device including a processor and a memory; the memory being used to store a computer program; and the processor being used to execute the steps of the method described in the first aspect by invoking the stored computer program.

[0024] A fifth aspect of the present invention provides a computer program product comprising a computer program and / or instructions that, when executed by a processor, implement the steps of the method as described in the first aspect.

[0025] In summary, this application has the following beneficial effects:

[0026] First, by adopting an external, non-intrusive architecture, the HSE intelligent agent attaches to the logged-in browser instance using a low-level debugging protocol, fully reusing the existing functions of the original HSE system. No modifications are required to the original HSE system, nor is it necessary to develop a new application interface or data platform. The implementation cost is low, the launch cycle is short, and it is seamlessly integrated with the enterprise's existing information system.

[0027] Second, by employing a three-tiered degradation strategy (API interception, DOM parsing, and offline file export) to collect data on potential hazards across systems, we ensure that data can be stably acquired under different system environments and technical conditions, thereby enhancing the robustness and coverage of data collection.

[0028] Third, by establishing cross-table association chains through key fields and using the rule engine to perform cross-validation, data scattered across multiple HSE systems such as non-routine operation systems, major hazard source systems, and personnel positioning systems are correlated and analyzed to automatically discover safety hazards across systems, such as abnormal situations where safety measures are not checked or alarms are overdue and not handled, thereby improving the accuracy and comprehensiveness of hazard discovery.

[0029] Fourth, a three-level fuzzy matching strategy (precise matching, fuzzy matching, and default matching) is used to automatically assign hidden danger tasks, ensuring that each hidden danger can be accurately assigned to the appropriate responsible unit, and eliminating unclear attribution, assignment delays, or assignment errors caused by manual assignment.

[0030] Fifth, the HSE intelligent agent automatically processes hazard data throughout the entire process, replacing manual operations and achieving standardized, traceable, and statistically accurate hazard handling, significantly reducing manual workload and eliminating the omission of hazards. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the security risk data processing method based on HSE intelligent agents according to the present invention.

[0032] Figure 2 This is a structural diagram of the safety hazard data processing system based on HSE intelligent agents according to the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0034] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0035] Example 1

[0036] This embodiment provides a security vulnerability data processing method based on an HSE agent, which is executed by the HSE agent and operates the HSE system using an external, non-intrusive architecture. Figure 1 Here is a flowchart illustrating the method, such as Figure 1 As shown, the method includes the following steps S1 to S5.

[0037] Step S1: Attach to the logged-in browser instance via the underlying debugging protocol and take over the control of the interaction between the browser instance and the HSE system.

[0038] Specifically, the HSE agent is deployed on the user-side terminal and attaches to the user's logged-in browser instance via a low-level debugging protocol (such as Chrome DevTools Protocol, CDP). Because it reuses the existing login session within the browser instance, the HSE agent can take over control of the interaction between the browser instance and the HSE system without needing to obtain or store the user's account password. This includes reading page response data, triggering page operation commands, and controlling file download paths. This plug-in, non-intrusive architecture allows the HSE agent to fully reuse the existing functionality of the original HSE system without requiring any interface modifications or secondary development of the original HSE system.

[0039] Step S2: Collect hidden danger operation data across systems using a three-level degradation strategy.

[0040] Specifically, the first layer is API interception: it prioritizes intercepting backend API responses and classifies and identifies hazardous operation data according to data structure characteristics (such as JSON array structure, pagination fields, etc.); if the SPA cache hit prevents the real request from being obtained, it forces a page reload instruction to retrieve the data again. The second layer is DOM parsing: if API interception fails, it returns to the page document and retrieves the hazardous operation data through DOM model parsing; in this process, it uses pre-defined multi-component library adaptation rules and multi-selector combinations to locate and extract page elements. The two component libraries are the Element UI component library and the Sinochem component library. Multiple candidate selectors are configured for the same type of page element, and the next candidate selector is automatically switched when a selector fails. The third layer is offline file export: when online retrieval of hazardous operation data fails, it controls the browser instance to force a specified download path, exports offline files (such as Excel reports), and parses the offline files to collect hazardous operation data across systems.

[0041] Step S3: Aggregate the hazard operation data obtained from multiple systems into the central database, establish cross-table association chains through key fields, and use the rule engine to perform cross-validation on the hazard operation data of multiple systems to obtain the validation status field.

[0042] Specifically, the work ticket number, source identifier, and company name are extracted as key fields from the non-routine operation system, major hazard source system, and personnel positioning system. These key fields across systems are cleaned (e.g., spaces are removed, encoding formats are standardized, and company names are standardized) and linked to establish cross-table association chains. The key fields (work ticket number, source identifier, and company name) are the connection nodes between the cross-system data. The cross-table association chain is a logical association formed by linking heterogeneous records belonging to the same operation in each system through these connection nodes. The rule engine is the main body for verification execution, loading verification rules onto the record groups corresponding to the same connection key in the association chain and outputting the verification status field. The rule engine performs cross-system cross-verification based on the cross-table association chain. For example, it compares the safety measure status in the work ticket with the actual alarm handling status. When a safety measure is detected as not involved, or an alarm is timed out and not handled, a safety hazard is determined, and the verification status field is set to a value indicating that the verification is successful; otherwise, it is set to a value indicating that the verification is unsuccessful.

[0043] Step S4: Determine whether to construct a fault work order corresponding to the hidden danger operation data in the browser instance based on the verification status field; if the verification status field indicates that the verification is successful, construct the fault work order and execute the three-level fuzzy matching strategy according to the preset attribution relationship to dispatch the fault work order to the responsible unit.

[0044] Specifically, when the verification status field indicates that the verification is successful, the HSE agent automatically constructs a fault work order in the browser instance, automatically filling in fields such as hazard description, hazard category, and department. Then, a three-level fuzzy matching strategy is executed: Level 1, precise matching is performed between the department field in the hazard operation data and a preset responsible unit mapping table; Level 2, when precise matching fails, fuzzy matching is performed between the hazard category field in the hazard operation data and preset business attribution rules; Level 3, when fuzzy matching fails, the fault work order is dispatched to the superior attribution unit according to preset default rules, ensuring that each fault work order has a clear responsibility attribution and no delays occur.

[0045] Step S5: Repeat steps S2 to S4 according to the preset cycle.

[0046] Specifically, the HSE intelligent agent automatically captures hazard operation data according to a preset cycle (e.g., once a day) and repeats steps S2 to S4 to achieve unattended periodic supervision. Furthermore, after a fault work order is dispatched to the responsible unit, the HSE intelligent agent periodically reviews the hazard rectification status corresponding to the fault work order according to a preset review cycle. When the review result is unsuccessful, the reviewed object is designated as the new work order responsible object, a new fault work order is generated and dispatched (e.g., escalated to the superior responsible unit). When the hazard operation data processing is detected as complete, the fault work order is automatically closed, thus forming an unattended data processing mechanism for safety hazards: "collection, verification, dispatch, review, and closure."

[0047] Example 2

[0048] This embodiment provides a security hazard data processing system based on HSE intelligent agents. Figure 2 This is a schematic diagram of the system structure. For example... Figure 2 As shown, the system adopts an external, non-intrusive architecture to operate the HSE system, including: a protocol attachment module, a degradation acquisition module, a data aggregation and verification module, a task construction and dispatch module, and a periodic scheduling module.

[0049] The protocol attachment module is used to attach to the logged-in browser instance via the underlying debugging protocol, taking over the interaction control between the browser instance and the HSE system. This module corresponds to the processing procedure in step S1 of Embodiment 1, which will not be described again here.

[0050] The degradation acquisition module is used to collect hazard operation data across systems using a three-layer degradation strategy. This module corresponds to the processing procedure in step S2 of Example 1, which will not be described again here.

[0051] The data aggregation and verification module is used to aggregate hazard operation data obtained from multiple systems to the central database. It establishes cross-table relationships through key fields and uses a rule engine to perform cross-validation on the hazard operation data from multiple systems to obtain the verification status field. This module corresponds to the processing procedure in step S3 of Example 1, and will not be described further here.

[0052] The task construction and dispatch module is used to determine whether to construct a fault work order corresponding to the hidden danger operation data in the browser instance based on the verification status field. When the verification status field indicates that the verification is successful, a three-level fuzzy matching strategy is executed according to the preset attribution relationship to dispatch the fault work order to the responsible unit. This module corresponds to the processing procedure in step S4 of embodiment one.

[0053] The periodic scheduling module is used to trigger the repeated execution of the degradation acquisition module, data aggregation and verification module, and task construction and dispatch module according to a preset period. This module corresponds to the processing procedure in step S5 of Embodiment 1.

[0054] Preferably, the system also includes a review and cancellation module, which is used to periodically review the rectification status of the hidden dangers corresponding to the fault work order according to a preset review cycle after the fault work order is dispatched to the responsible unit; when the review result is not passed, the reviewed object is taken as the new work order responsible object, a new fault work order is generated and dispatched; when the hidden danger operation data is detected to be completed, the fault work order is automatically cancelled.

[0055] Example 3

[0056] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0057] This embodiment also provides an electronic device, including a processor and a memory; the memory is used to store a computer program; the processor is used to call the computer program stored in the memory and execute the steps of the method as described in Embodiment 1.

[0058] This embodiment also provides a computer program product, including a computer program and / or instructions, which, when executed by a processor, implement the steps of the method as described in Embodiment 1.

[0059] The above description is merely 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 within the scope of protection of the present invention.

Claims

1. A method for processing safety hazard data based on HSE intelligent agents, characterized in that, The HSE system is operated using an external, non-intrusive architecture, and the method is executed by the HSE agent, including the following steps: S1. Attach to the logged-in browser instance through the underlying debugging protocol and take over the control of interaction with the HSE system; S2. Use a three-tier degradation strategy to collect hidden danger operation data across systems; S3. Aggregate the hidden danger operation data obtained from the cross-system to the central database, establish cross-table association chains through key fields, and use the rule engine to perform cross-validation on the hidden danger operation data of multiple systems based on the cross-table association chains to obtain the validation status field. S4. Determine whether to construct a fault work order for the hidden danger operation data in the browser instance based on the verification status field; if the verification status field is valid, then execute a three-level fuzzy matching strategy according to the preset attribution relationship to dispatch the fault work order to the responsible unit. S5. Repeat steps S2-S4 according to the preset cycle.

2. The method according to claim 1, characterized in that, The three-tier degradation strategy includes: prioritizing the interception of backend API responses and classifying and identifying them according to data structure characteristics; if the SPA cache hit causes the inability to obtain the real request, then forcibly triggering the page reload instruction to obtain it. If the retrieval fails, return to the page document and retrieve the hazard operation data through DOM model parsing; When online acquisition of the hazard operation data fails, the browser instance is controlled to force the specified download path to export offline files and parse them, thereby collecting hazard operation data across systems.

3. The method according to claim 2, characterized in that, The returned page document obtains the hidden danger operation data through DOM model parsing, including: locating and extracting page elements through a combination of pre-set multi-component library adaptation rules and multi-selectors, wherein the multi-component library adaptation rules are used to ensure compatibility with different front-end UI component libraries.

4. The method according to claim 1, characterized in that, The three-level fuzzy matching strategy includes: performing precise matching based on the department field in the hidden danger operation data and a preset responsibility unit mapping table; when precise matching fails, performing fuzzy matching based on the hidden danger category field in the hidden danger operation data and a preset business affiliation rule. When fuzzy matching fails, the fault work order corresponding to the hidden danger operation data is dispatched to the superior unit according to the preset default rules.

5. The method according to claim 1, characterized in that, The step of establishing a cross-table relationship chain through key fields and using a rule engine to perform cross-validation on the hidden danger operation data of multiple systems based on the cross-table relationship chain includes: Extract the work ticket number, source identifier, and company name from the non-routine operation system, major hazard source system, and personnel positioning system as the key fields; By collecting the key fields and cleaning and associating the fields across systems, the cross-table association chain is established; The rule engine performs cross-system cross-validation based on cross-table association chains. The cross-validation includes comparing the status of safety measures in the work order with the actual alarm handling status. When it is detected that a safety measure is selected as not involved or the alarm has expired and not been handled, it is determined that there is a safety hazard.

6. The method according to claim 1, characterized in that, The method also includes, After the fault work order is dispatched to the responsible unit, the rectification status of the hidden dangers corresponding to the fault work order is reviewed regularly according to the preset review cycle. When the review result is not passed, the reviewed object will be treated as the new work order responsible object, a new fault work order will be generated and dispatched; When the processing of the hazard operation data is completed, the fault work order is automatically closed.

7. A safety hazard data processing system based on HSE intelligent agents, characterized in that, The HSE system operates using an external, non-intrusive architecture, and the system includes: The protocol attachment module is used to attach to the logged-in browser instance via the underlying debugging protocol and take over the control of interaction with the HSE system; The degradation acquisition module is used to collect hazard operation data across systems using a three-layer degradation strategy; The data aggregation and verification module is used to aggregate the hidden danger operation data obtained from cross systems to the central database, establish cross-table association chains through key fields, and use the rule engine to perform cross-verification on the hidden danger operation data of multiple systems based on the cross-table association chains to obtain the verification status field. The task construction and dispatch module is used to determine whether to construct the fault work order corresponding to the hidden danger data in the browser instance based on the verification status field, and dispatch the fault work order corresponding to the hidden danger operation data to the responsible unit according to the preset attribution relationship and the three-level fuzzy matching strategy. The periodic scheduling module is used to trigger the repeated execution of the degradation acquisition module, data aggregation and verification module, and task construction and dispatch module according to a preset period.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method of any one of claims 1 to 6.

9. An electronic device, characterized in that, Including processor and memory; The memory is used to store computer programs; The processor is configured to execute the method of any one of claims 1 to 6 by invoking a stored computer program.

10. A computer program product comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 6.

Images