An intelligent management and control system for safety risks in a chemical industrial park

Abstract

The application relates to an intelligent safety risk management and control system for a chemical industrial park, which comprises the following modules: a risk identification and evaluation module, which acquires park planning layout information, real-time operation data of a public pipe gallery and production process data of associated enterprises, identifies key device risk items of the associated enterprises caused by supply interruption of the public pipe gallery based on a pre-set risk identification model, and outputs a risk item list and associated influence paths of each risk item; a relay storage planning module, which calculates the minimum relay storage capacity required by each risk item and plans to generate a relay storage scheme; a relay storage management module, which executes each relay storage scheme and enables emergency switching when the public pipe gallery fails; and a pipe gallery monitoring and early warning module, which collects the operation state of the public pipe gallery and triggers emergency switching when the public pipe gallery is in an abnormal state. The application significantly reduces the scale of chain accidents caused by pipe gallery failure in the chemical industrial park and improves the systematic prevention and control capability of safety risks in the chemical industrial park.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent management and control, in particular to a chemical industrial park safety risk intelligent management and control system. BACKGROUND

[0002] A chemical industrial park refers to a specific area in which chemical enterprises are concentrated and arranged within a certain geographical range, and through scientific planning and reasonable matching, resource sharing, complementary advantages and coordinated development are achieved. As an important carrier for the development of the petroleum and chemical industry, the chemical industrial park has played a strong leading and supporting role in promoting the high-end, differentiation and green development of the petrochemical industry and improving the core competitiveness of the industry.

[0003] The chemical industrial park has a high degree of spatial agglomeration of enterprises and a strong industrial chain correlation. A single accident can quickly spread to the surrounding area, triggering a domino effect and forming a "chain reaction". When planning a chemical industrial park, the industrial types of enterprises in the park, production needs and industrial chain correlations are often considered to plan a public pipe gallery. Compared to a single enterprise accident, a problem in the public pipe gallery can trigger a more serious and difficult-to-control chain reaction. Due to the complexity of the transported medium (mostly flammable, explosive, toxic and hazardous chemical materials) and the closed space, if a section of the pipe leaks or explodes, it can disrupt the entire park's material supply and even cause the reaction kettle of upstream and downstream enterprises to lose control and products to accumulate. At this time, the domino effect caused is often strong, and the scope of the accident is geometrically enlarged.

[0004] To better deal with the chain reaction when a problem occurs in the public pipe gallery, a chemical industrial park safety risk intelligent management and control system is needed. SUMMARY

[0005] To solve the above problems, the present application provides a chemical industrial park safety risk intelligent management and control system.

[0006] In a first aspect, the present application provides a chemical industrial park safety risk intelligent management and control system, which adopts the following technical solution:

[0007] A chemical industrial park safety risk intelligent management and control system, comprising:

[0008] A risk identification and assessment module for obtaining park planning layout information, public pipe gallery real-time operation data and associated enterprise production process data, identifying key device risk items of associated enterprises caused by public pipe gallery supply interruption based on a pre-set risk identification model, and outputting a risk item list and associated influence paths of each risk item; the risk identification model is a machine learning model obtained by feature engineering processing and iterative training on public pipe gallery historical accident data, park layout data and enterprise production process data;

[0009] a relay storage planning module connected to the risk identification and assessment module, configured to calculate the minimum relay storage capacity required by each risk item according to the safety interruption time window T of the risk item and the medium consumption Q of single batch production, and to plan and generate a relay storage scheme including relay storage type information, recommended storage capacity information and access path information; wherein the safety interruption time window T of the risk item is the critical time from the public pipeline medium interruption to the triggering of the risk item output by the risk identification model;

[0010] a relay storage management module configured to execute each relay storage scheme, to schedule the public pipeline to fill the relay storage with the medium consumption Q of single batch production at the beginning of the production cycle of the associated enterprise, to call the relay storage for feeding at the last batch of the production cycle, and to enable emergency switching when the public pipeline fails;

[0011] a pipeline monitoring and early warning module configured to collect the running state of the public pipeline and trigger the emergency switching of the relay storage management module when the public pipeline is in an abnormal state.

[0012] Preferably, the risk identification and assessment module comprises:

[0013] an information acquisition unit configured to collect and acquire park planning layout information, public pipeline real-time running data and production process data of associated enterprises; the park planning layout information includes park GIS geographic information, public pipeline topology information, park building layout information and enterprise key device layout information; the public pipeline real-time running data includes medium attribute information and state information of each public pipeline transportation and connection enterprise information;

[0014] an influence path analysis unit configured to construct a supply dependence network of public pipeline-enterprise-production device, and to identify the key path triggering a chain reaction when the public pipeline failure causes medium interruption;

[0015] a risk identification unit configured to identify the risk items of key devices of associated enterprises caused by public pipeline supply interruption based on a pre-set risk identification model, and to output a risk item list and associated influence paths of each risk item, the risk item list including public pipeline number, enterprise name, key device ID, risk type, safety interruption time window T, required medium, and minimum maintenance flow Pmin;

[0016] a path visualization unit configured to highlight the associated influence paths in the dependence network, and to output an interactive topological graph and path description text, the associated influence paths being the influence paths from the public pipeline failure point to the key devices at risk.

[0017] Preferably, the dependence network further comprises a secondary dependence network based on the direct supply of production device output to secondary production lines.

[0018] Preferably, the relay storage planning module comprises:

[0019] A relay storage capacity evaluation unit is configured to calculate the minimum capacity of the relay storage required by each risk item according to the safety interruption time window T of the risk item and the medium consumption Q of single batch production, through a pre-set minimum relay storage capacity calculation formula, and obtain the corresponding safety factor of the minimum capacity according to a pre-set capacity coefficient table, and the product of the minimum capacity and the safety factor is the recommended storage capacity of the relay storage required by the risk item.

[0020] A scheme planning unit is configured to generate a relay storage scheme by a pre-set scheme planning model, which is a machine learning model obtained by iterative training of historical relay storage scheme data and historical risk accident analysis scheme data, wherein the historical risk accident analysis scheme data is the corresponding relay storage scheme data given by an expert team according to historical risk events.

[0021] Preferably, the minimum relay storage capacity calculation formula is specifically:

[0022] Vmin=max(T×Pmin, Q).

[0023] Preferably, the relay storage management module comprises:

[0024] A production cycle synchronization unit is configured to obtain a production plan by connecting an enterprise manufacturing execution system, trigger a medium filling instruction of the common pipe gallery to the relay storage at a production cycle starting node, and the filling amount is the single batch consumption Q.

[0025] A production switching unit is configured to call the relay storage to supply the production line at the last batch of the production cycle, and trigger an emergency switching instruction when the pipe gallery monitoring and early warning module sends a fault signal, cut off the common pipe gallery supply management, and synchronously call the relay storage to supply the production line.

[0026] Preferably, the relay storage management module further comprises:

[0027] A medium evaluation unit is configured to determine whether the remaining effective period of the storage medium in the relay storage is less than a pre-set effective period threshold at the last batch of the production cycle according to the production plan, and if so, issue a replacement instruction to the production switching unit when the remaining effective period of the storage medium in the relay storage reaches the effective period threshold, switch to call the relay storage to supply the production line at the next batch of production, and after the supply of this batch of production is completed, pause the production, trigger the medium filling instruction of the common pipe gallery to the relay storage again, the filling amount is the single batch consumption Q, and control the common pipe gallery to continue supplying the production line for continuous production after the filling is completed.

[0028] Preferably, the production switching unit performs an inert gas replacement and a medium residual amount detection procedure after a batch production is completed by calling the relay storage to supply the production line.

[0029] Preferably, the information acquisition unit is connected with a chemical industrial park in-out cargo information management terminal, and periodically acquires in-out cargo lists of each enterprise to verify whether the production process changes, and if so, sends a process change request to the enterprise manager.

[0030] Preferably, the periodic acquisition of in-out cargo lists of each enterprise to verify whether the production process changes specifically includes: based on the production process data of the enterprise, periodically acquiring in-out cargo lists of each enterprise to generate an enterprise actual material set A and an enterprise recorded material set B, calculating the Jaccard similarity S of the actual in-out material and the recorded process material through a pre-set consistency verification formula, and if the Jaccard similarity S is less than a pre-set similarity threshold J, it is determined that the production process of the enterprise changes; wherein J is pre-set by the manager, and 0.7≤J≤0.9; the consistency verification formula is specifically: .

[0031] In summary, the present application includes at least one of the following beneficial technical effects:

[0032] 1. According to the actual situation of the chemical industrial park, the risk identification model identifies and determines the key device risk items of the associated enterprises caused by the interruption of the public pipe gallery supply, which can accurately identify the key nodes of the "domino effect" that may be triggered after the pipe gallery interruption; and then configure a relay storage scheme for each risk item and set a relay storage to ensure continuous supply of media when the public pipe gallery stops supplying due to failure, which not only avoids the chain risk caused by the failure of the public pipe gallery in the chemical industrial park, but also realizes dynamic resource optimization, miniaturizes the relay storage, reduces the cost of risk control facilities, avoids new hidden danger points, and transforms the domino effect of the public pipe gallery accident from uncontrollable geometric expansion to controllable single-point blocking, significantly reduces the scale of chain accidents caused by pipe gallery failure in the chemical industrial park, significantly improves the systematic prevention and control ability of safety risks in the chemical industrial park, and achieves the effect of effective and efficient comprehensive prevention and control of the chemical industrial park.

[0033] 2. The information acquisition unit acquires four core data categories: spatial geographic data of the chemical park, facility topology data, medium operation data, and production process data. The impact path analysis unit then constructs a supply dependency network of the public pipeline corridor, enterprise, and production equipment, and further constructs a secondary dependency network. This not only identifies the direct impact of pipeline corridor accidents, but also penetrates into the secondary production links, accurately locating the transmission chain of the domino effect. This helps the risk identification model accurately and efficiently identify the key device risk items of related enterprises caused by public pipeline corridor supply interruptions, facilitates subsequent efficient and accurate relay storage planning, and significantly enhances the systematic prevention and control capabilities of chemical park safety risks. The path visualization unit then highlights the associated impact paths in the dependency network and outputs interactive topology maps and path description text, allowing managers to clearly understand the risk impact, significantly improving the management's intervention response speed to those minor risk impacts, and effectively curbing the expansion of the domino effect.

[0034] 3. By comparing the Jaccard similarity between the company's actual inbound and outbound cargo lists and the registered material collection, unregistered process changes (such as new raw materials or by-products) can be quantitatively identified, avoiding risk assessment failures caused by hidden process changes and solving the problems of low efficiency and high missed detection rates in traditional manual verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a system block diagram of an intelligent safety risk management and control system for a chemical park in an embodiment of the present application;

[0036] Figure 2 This is a system block diagram of the risk identification and assessment module in an embodiment of the present application;

[0037] Figure 3 This is a system block diagram of the relay storage planning module in an embodiment of the present application;

[0038] Figure 4 It is a system block diagram of the relay storage management module in an embodiment of the present application.

[0039] Explanation of the accompanying symbols: 1. Risk identification and assessment module; 11. Information acquisition unit; 12. Impact path analysis unit; 13. Risk identification unit; 14. Path visualization unit; 2. Relay storage planning module; 21. Relay storage capacity assessment unit; 22. Scheme planning unit; 3. Relay storage management module; 31. Production cycle synchronization unit; 32. Production switching unit; 33. Medium assessment unit; 4. Pipeline corridor monitoring and early warning module. DETAILED DESCRIPTION

[0040] The following combination Figures 1-4 This application is described in further detail.

[0041] The embodiment of the application discloses a chemical industrial park safety risk intelligent management and control system. Figure 1 The chemical industrial park safety risk intelligent management and control system comprises a risk identification and evaluation module 1, a relay storage planning module 2, a relay storage management module 3 and a pipe gallery monitoring and early warning module 4. The risk identification and evaluation module 1, the relay storage planning module 2, the relay storage management module 3 and the pipe gallery monitoring and early warning module 4 are connected in communication with each other. The risk identification and evaluation module 1 is used for acquiring park planning layout information, public pipe gallery real-time operation data and associated enterprise production process data, identifying key device risk items of the associated enterprise caused by public pipe gallery supply interruption based on a pre-set risk identification model, and outputting a risk item list and associated influence paths of each risk item. The risk identification model is a machine learning model obtained by performing feature engineering processing and iterative training on public pipe gallery historical accident data, park layout data and enterprise production process data. It should be noted that the specific training steps of the machine learning model are prior art, and will not be repeated here. The relay storage planning module 2 is connected to the risk identification and evaluation module 1 and is used for calculating the minimum relay storage capacity required by each risk item according to a safety interruption time window T of the risk item and a single batch production medium consumption Q, and planning and generating a relay storage scheme. The relay storage scheme comprises relay storage type information, recommended storage capacity information and access path information. The safety interruption time window T of the risk item is the critical time from public pipe gallery medium interruption to risk item triggering output by the risk identification model. The relay storage management module 3 is used for executing each relay storage scheme, scheduling public pipe gallery to fill the single batch production medium consumption Q into the relay storage at the beginning of the associated enterprise production cycle, calling the relay storage for feeding at the last batch of the production cycle, and enabling emergency switching when the public pipe gallery fails. The pipe gallery monitoring and early warning module 4 is used for collecting and monitoring the operating state of the public pipe gallery and triggering the emergency switching of the relay storage management module 3 when the public pipe gallery is in an abnormal state. According to the actual situation of the chemical industrial park, the risk identification model identifies and determines the key device risk items of the associated enterprise caused by public pipe gallery supply interruption, can accurately identify the key nodes of the “domino effect” that may be triggered after the pipe gallery interruption, and further configures a relay storage scheme for each risk item, sets a relay storage, and quickly switches the relay storage to ensure continuous medium supply when the public pipe gallery stops feeding due to failure. This not only avoids the chain risk caused by the failure of the public pipe gallery in the chemical industrial park, but also realizes dynamic resource optimization, miniaturizes the relay storage, reduces the cost of risk control facilities, avoids new hidden danger points, converts the domino effect of the public pipe gallery accident from uncontrollable geometric expansion to controllable single-point blocking, significantly reduces the scale of chain accidents caused by pipe gallery failure in the chemical industrial park, significantly improves the systematic prevention and control capability of the safety risk of the chemical industrial park, and achieves the effect of effective and efficient comprehensive prevention and control of the chemical industrial park.

[0042] Referring toFigure 2 The risk identification and assessment module 1 comprises:

[0043] An information acquisition unit 11 is configured to acquire park planning layout information, public pipe gallery real-time operation data and associated enterprise production process data; the park planning layout information comprises park GIS geographic information, public pipe gallery topology information, park building layout information and enterprise key device layout information; the public pipe gallery real-time operation data comprises medium attribute information and state information of each public pipe gallery and connection enterprise information;

[0044] An influence path analysis unit 12 is configured to construct a supply dependency network of the public pipe gallery-enterprise-production device, and identify a key path triggered by a chain reaction when the public pipe gallery failure causes medium interruption; in addition, the dependency network further comprises a secondary dependency network based on a direct supply of production device output to a secondary production line;

[0045] A risk identification unit 13 is configured to identify a key device risk item of an associated enterprise caused by public pipe gallery supply interruption based on a pre-set risk identification model, and output a risk item list and an associated influence path of each risk item, wherein the risk item list comprises a public pipe gallery number, an enterprise name, a key device ID, a risk type, a safety interruption time window T, a required medium and a minimum maintenance flow Pmin;

[0046] A path visualization unit 14 is configured to highlight the associated influence path in the dependency network, and output an interactive topological graph and path description text (such as “pipe gallery G3 leakage→influence enterprise E2→triggered reaction kettle R5 overpressure→may affect secondary production line L7”), and the associated influence path is an influence path from a public pipe gallery failure point to a key device at risk. By acquiring four categories of core data, i.e. chemical industry park spatial geographic data, facility topology data, medium operation data and production process data through the information acquisition unit 11, and then constructing a supply dependency network of the public pipe gallery-enterprise-production device through the influence path analysis unit 12, and further constructing a secondary dependency network, not only the direct influence of the pipe gallery accident can be identified, but also the transmission chain of the domino effect can be accurately located, which helps the risk identification model to accurately and efficiently identify the key device risk item of the associated enterprise caused by the public pipe gallery supply interruption, and helps to subsequently perform relay storage planning efficiently and accurately, thereby significantly improving the systematic prevention and control capability of the safety risk of the chemical industry park; and then the path visualization unit 14 highlights the associated influence path in the dependency network, and outputs an interactive topological graph and path description text, so that the management personnel can clearly understand the risk influence, significantly improve the intervention response speed of the management personnel to the small risk influence, and effectively curb the expansion of the domino effect.

[0047] Take the chain of "pipe gallery G3 leakage → affect enterprise E2 → trigger reaction kettle R5 overpressure → possibly affect secondary production line L7" as an example, when no relay memory is set, the reaction kettle R5 overpressure explodes, and the secondary production line L7 stops production or the reaction is intense because the reaction kettle feed is stopped; and after the relay memory is set, the relay memory switches the feed in response to the warning, and the reaction kettle runs stably, but there may be fluctuations in output, but this has little effect on the secondary production line L7. Management personnel can quickly locate the affected production line through the topological graph and intervene in production, such as activating standby raw materials.

[0048] In addition, the information acquisition unit 11 is also connected with a chemical industry park access cargo information management terminal, and periodically acquires the access cargo list of each enterprise to verify whether the production process changes, and if so, sends a process change request to the enterprise manager. By verifying the production process of each enterprise through the access cargo list, the process of the enterprise can be simply and efficiently identified and verified, forcing the enterprise to report process adjustments and ensuring that the park risk database is synchronized with the actual production state of the enterprise in real time, thereby reducing safety hazards caused by opaque processes from the source.

[0049] The periodically acquiring the access cargo list of each enterprise to verify whether the production process changes specifically includes: based on the production process data of the enterprise, periodically acquiring the access cargo list of each enterprise to generate an enterprise actual material set A and an enterprise recorded material set B, calculating the Jaccard similarity S of the actual access material and the recorded process material through a pre-set consistency verification formula, and if the Jaccard similarity S is less than a pre-set similarity threshold J, it is determined that the production process of the enterprise changes; wherein J is pre-set by the manager, and 0.7≤J≤0.9, and in the embodiment, the similarity threshold J is set to 0.8; the consistency verification formula is specifically: By comparing the Jaccard similarity of the actual access cargo list of the enterprise and the recorded material set, unrecorded process changes (such as adding raw materials or byproducts) can be quantitatively identified, avoiding risk assessment failure caused by hidden process changes, and solving the problems of low efficiency and high omission rate of traditional manual verification.

[0050] Referring to Figure 3 , the relay storage planning module 2 includes:

[0051] The relay storage capacity evaluation unit 21 is configured to calculate the minimum capacity of the relay storage required by each risk item according to the safe interruption time window T of the risk item and the medium consumption Q of single batch production, through a pre-set minimum relay storage capacity calculation formula, and obtain the corresponding safety factor of the minimum capacity according to a pre-set capacity factor reference table, and the product of the minimum capacity and the safety factor is the recommended storage capacity of the relay storage required by the risk item; in the capacity factor reference table, the larger the capacity, the smaller the corresponding safety factor, and the safety factor is greater than 1 and less than 1.3, so as to realize the safety factor step by step, and realize the storage strategy of “planning on demand and safe and economic”;

[0052] The scheme planning unit 22 is configured to generate a relay storage scheme through a pre-set scheme planning model, and the scheme planning model is a machine learning model obtained by iterative training of historical relay storage scheme data and historical risk accident analysis scheme data, wherein the historical risk accident analysis scheme data is corresponding relay storage scheme data given by an expert team according to historical risk events. The specific training steps of the machine learning model are prior art, and are not described here. Through the relay storage capacity evaluation unit 21, the minimum capacity of the relay storage is calculated and determined according to the safe interruption time window T of the risk item and the medium consumption Q of single batch production, and the recommended capacity is dynamically adjusted in combination with the safety factor, so as to avoid risks caused by insufficient storage and prevent resource waste caused by excess capacity, realize dynamic resource optimization, miniaturize the relay storage, reduce the cost of risk control facilities, and avoid new hidden danger points; and then the scheme planning model efficiently matches the optimal storage scheme for each risk item according to the actual demand situation such as the scene, capacity and medium of the risk item.

[0053] The minimum relay storage capacity calculation formula is specifically as follows:

[0054] Vmin=max(T×Pmin, Q). Through the formula, three contradictions are balanced: safety must be 100% guaranteed, but resources are always limited, and production cannot be stopped. T×Pmin ensures that there is enough buffer to maintain safe shutdown of the device when the pipe gallery is interrupted, Q guarantees the completion of single batch production, avoids imbalance of materials caused by interruption in the middle of the way, and effectively suppresses the expansion of the domino effect.

[0055] Referring to Figure 4 , the relay storage management module 3 comprises:

[0056] The production cycle synchronization unit 31 is configured to obtain a production plan by connecting an enterprise manufacturing execution system, trigger a medium filling instruction of the common pipe gallery to the relay storage at a production cycle starting node, and the filling amount is the single batch consumption Q;

[0057] The production switching unit 32 is used to call the relay storage to supply the production line at the last batch of the production cycle, and trigger the emergency switching instruction when the pipeline monitoring and early warning module 4 sends a fault signal, cut off the public pipeline supply management, and synchronously call the relay storage to supply the production line;

[0058] The medium evaluation unit 33 is used to determine whether the remaining validity period of the medium stored in the relay storage is less than the preset validity threshold at the last batch of the production cycle according to the production plan, and if so, send a replacement instruction to the production switching unit 32 when the remaining validity period of the medium stored in the relay storage reaches the validity threshold, switch to call the relay storage to supply the production line at the next batch of production, and after the supply of this batch of production is completed, suspend the production, trigger the medium filling instruction of the public pipeline to the relay storage again, and the filling amount is the single batch consumption Q, and after the filling is completed, control the public pipeline to continue to supply the production line to continue the production. Through the connection of the production cycle synchronization unit 31 to the enterprise manufacturing execution system (MES), the medium filling and switching are automatically triggered according to the production plan, so as to avoid the time error caused by manual intervention, ensure that the storage medium is "ready to use" when a fault occurs, and improve the timeliness of emergency response.

[0059] In addition, the production switching unit 32 executes the inert gas replacement and medium residual amount detection program after calling the relay storage to supply the production line to complete a batch of production. Through the medium evaluation unit 33, the validity period of the storage medium is monitored in real time, and combined with the replacement instruction and the inert gas replacement program, the influence of expired medium on production safety is prevented, and through the residual amount detection, the risk of chemical reaction caused by mixing of different media is avoided, and the whole-process safety control of storage management is realized.

[0060] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the protection scope of the application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art can still combine, add or delete the features of the embodiments of the present application according to the circumstances without creative labor, so as to obtain different other technical solutions which do not deviate from the concept of the present application in essence, and these technical solutions also belong to the scope of protection of the present application.

Claims

1. An intelligent safety risk management and control system for a chemical park, characterized by: include: The risk identification and assessment module (1) is used to obtain park planning and layout information, real-time operation data of the public pipeline corridor and production process data of related enterprises, identify key device risk items of related enterprises caused by supply interruption of the public pipeline corridor based on a preset risk identification model, and output a risk item list and the associated impact path of each risk item; the risk identification model is a machine learning model obtained by feature engineering processing and iterative training of historical accident data of the public pipeline corridor, park layout data and enterprise production process data, and the risk item list includes the public pipeline corridor number, enterprise name, key device ID, risk type, safe interruption time window T, required medium, and minimum maintenance flow Pmin; The relay storage planning module (2) is connected to the risk identification and assessment module (1) and is used to calculate the minimum relay storage capacity required for each risk item based on the safe interruption time window T of the risk item and the medium consumption Q of a single batch of production, and plan and generate a relay storage plan, wherein the relay storage plan includes relay storage type information, recommended storage capacity information, and access path information; wherein the safe interruption time window T of the risk item is the critical time from the interruption of the public pipeline corridor medium to the triggering of the risk item output by the risk identification model; The relay storage management module (3) is used to execute various relay storage schemes, dispatch the public pipeline corridor to fill the relay storage with the medium consumption Q of a single batch of production at the beginning of the production cycle of the associated enterprise, call the relay storage for material supply at the last batch of the production cycle, and enable emergency switching when the public pipeline corridor fails; A pipe gallery monitoring and early warning module (4) is used to collect and monitor the operating status of the public pipe gallery and trigger the emergency switching of the relay storage management module (3) when the public pipe gallery is in an abnormal state; The relay storage planning module (2) comprises: The relay storage capacity evaluation unit (21) is used to calculate the minimum capacity of the relay storage required for each risk item according to the safety interruption time window T of the risk item and the medium consumption Q of a single batch of production using a preset minimum relay storage capacity calculation formula, and obtain the corresponding safety factor of the minimum capacity according to a preset capacity factor comparison table, and the product of the minimum capacity and the safety factor is the recommended storage capacity of the relay storage required for the risk item; A scheme planning unit (22) is used to generate a relay storage scheme through a preset scheme planning model, wherein the scheme planning model is a machine learning model obtained by iteratively training historical relay storage scheme data and historical risk accident analysis scheme data, wherein the historical risk accident analysis scheme data is corresponding relay storage scheme data given by an expert team based on historical risk events; The calculation formula for the minimum relay storage capacity is specifically: Vmin=max(T×Pmin,Q).

2. The intelligent safety risk management and control system for a chemical park according to claim 1 is characterized in that: The risk identification and assessment module (1) includes: An information acquisition unit (11) is used to collect and acquire park planning and layout information, real-time operation data of public pipeline corridors, and production process data of related enterprises; the park planning and layout information includes park GIS geographic information, public pipeline corridor topology information, park building layout information, and enterprise key device layout information; the real-time operation data of public pipeline corridors includes medium attribute information and status information of each public pipeline corridor and connected enterprise information; An impact path analysis unit (12) is used to construct a supply dependency network of a public pipeline corridor, an enterprise, and a production device, and to identify a critical path that triggers a chain reaction when a public pipeline corridor failure causes a medium interruption; A risk identification unit (13) is used to identify key device risk items of associated enterprises caused by the interruption of public pipeline supply based on a preset risk identification model, and output a list of risk items and associated impact paths of each risk item; The path visualization unit (14) is used to highlight the associated impact path in the dependent network and output an interactive topology map and path description text, where the associated impact path is the impact path from the public pipeline corridor fault point to the key device with risk.

3. The intelligent safety risk management and control system for a chemical park according to claim 2 is characterized by: The dependency network also includes a secondary dependency network based on the output of the production device directly supplying the secondary production line.

4. The intelligent safety risk management and control system for a chemical park according to claim 1 is characterized in that: The relay storage management module (3) comprises: A production cycle synchronization unit (31) is used to connect to the enterprise manufacturing execution system to obtain a production plan, and trigger a medium filling instruction from the public pipe gallery to the relay storage at the start node of the production cycle, with the filling amount being a single batch consumption Q; The production switching unit (32) is used to call the relay storage to feed the production line in the last batch of the production cycle, and trigger the emergency switching instruction when the corridor monitoring and early warning module (4) sends a fault signal, cut off the public corridor feeding management, and synchronously call the relay storage to feed the production line.

5. The intelligent safety risk management and control system for a chemical park according to claim 4 is characterized in that: The relay storage management module (3) further includes: The medium evaluation unit (33) is used to judge whether the remaining validity period of the storage medium in the relay memory is less than a preset validity period threshold value when the last batch of the production cycle is judged according to the production plan. If it is less, when the remaining validity period of the storage medium in the relay memory reaches the validity period threshold value, a material change instruction is sent to the production switching unit (32), and the relay memory is switched to supply the production line during the next batch of production. After the supply of the production of this batch is completed, the production is suspended, and the public pipeline is triggered to refill the medium to the relay memory. The filling amount is the single batch consumption Q. After the filling is completed, the public pipeline is controlled to continue to supply the production line to continue production.

6. The intelligent safety risk management and control system for a chemical park according to claim 5 is characterized by: The production switching unit (32) executes an inert gas replacement and medium residual amount detection program after calling the relay memory to supply materials to the production line to complete a batch of production.

7. The intelligent safety risk management and control system for a chemical park according to claim 2 is characterized by: The information acquisition unit (11) is connected to the inbound and outbound goods information management terminal of the chemical park, and regularly obtains the inbound and outbound goods lists of each enterprise to verify whether there are changes in their production processes. If so, a process change request is sent to the enterprise management personnel.

8. The intelligent safety risk management and control system for a chemical park according to claim 7, characterized in that: The method of regularly obtaining the inbound and outbound cargo lists of each enterprise to verify whether there are changes in its production process specifically includes: based on the production process data of the enterprise, regularly obtaining the inbound and outbound cargo lists of each enterprise to generate the enterprise's actual material set A and the enterprise's registered material set B, and calculating the Jaccard similarity S between the actual inbound and outbound materials and the registered process materials using a preset consistency verification formula. If the Jaccard similarity S is less than a preset similarity threshold J, it is determined that there are changes in the enterprise's production process; wherein J is preset by the administrator and 0.7≤J≤0.9; the consistency verification formula is specifically: .

Images