A method for dynamic identity and permission adaptive management under a zero-trust architecture
By converting the physical address of resource objects into the fire-affected address under a zero-trust architecture, and combining this with personnel and equipment permission verification, the problem of permission mismatch in fire emergency scenarios is solved, enabling dynamic permission adjustment and precise control.
Patent Information
- Application Number
- CN202611097635.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-23
- Publication Date
- 2026-11-17
AI Technical Summary
Existing zero-trust access control solutions are difficult to adapt to changes in resource and object risk attribution caused by boundary migration due to fire impact in industrial fire emergency scenarios. This can lead to situations where the access personnel are legitimate but their access behavior does not match the current fire risk, resulting in miscontrol or linkage conflicts.
By converting the original physical address of the target resource object into the target fire-affected address, and combining the current identity verification result of the access personnel with the command permissions on the device, dynamic permission adjustments are made to ensure that the permission judgment basis changes from static physical attribution to dynamic risk attribution, thus realizing dual-end permission verification.
It improves the accuracy of access control, reflects the true safety implications of resource objects in fire scenarios in real time, avoids the problem of access object mismatch caused by the migration of fire scene boundaries, and enhances the security and accuracy of control commands.
Smart Images

Figure CN123122871A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of network security and identity and access control technology, and in particular to a dynamic identity and access control method under a zero-trust architecture. Background Technology
[0002] In a zero-trust security architecture, every access request requires re-verification of identity status, access environment, task association, and resource attributes. However, in emergency scenarios such as industrial fires, access control faces unique challenges. The intensity of fire, smoke, and heat radiation change over time, causing data, equipment, and other resources on the platform to have different security implications at different times. For example, a smoke exhaust fan may physically belong to workshop A, but when smoke affects the evacuation route of workshop B, the fan becomes a critical object affecting the evacuation safety of workshop B. If authorization is still granted according to workshop A's permissions, it could lead to workers in workshop A making incorrect operations despite their legitimate identities, thus impacting the safety of workshop B.
[0003] Existing zero-trust access control solutions typically assume that the ownership of resource objects is fixed, setting access policies around this fixed ownership. However, after a fire occurs, the security implications of resources change due to the spread of the fire, and fixed ownership cannot reflect the current risk attribution of resources. Furthermore, existing solutions usually adjust permissions based on the identity of the accessing personnel, but even if a person's identity is legitimate, the object they request access to may become unsuitable for access due to the shift of fire boundaries. Simultaneously, existing solutions treat equipment as passively controlled resources, only determining whether the accessing personnel have the right to control it, without redefining the authorized boundaries of the equipment itself within the current fire spread context.
[0004] To address the above issues, this application proposes a dynamic identity and permission adaptive management method under a zero-trust architecture. Summary of the Invention
[0005] The technical problem this invention aims to solve is that, in the emergency response to industrial fires, existing zero-trust access control methods typically determine access permissions based on static resource ownership relationships. This makes it difficult to adapt in a timely manner to changes in the risk ownership of resource objects caused by the migration of boundaries due to the fire. This can easily lead to situations where the identity of the person accessing the fire is legitimate, but their access behavior does not match the current fire risk, resulting in miscontrol or linkage conflicts.
[0006] To address the aforementioned technical problems, this invention provides a dynamic identity and permission adaptive control method under a zero-trust architecture. The method responds to access requests initiated by users for target resource objects, and includes:
[0007] The original physical address of the target resource object is converted into the target fire impact address, wherein the target fire impact address is determined based on the current fire propagation front and is used to characterize the ownership of the target resource object under the current fire propagation relationship;
[0008] Based on the current identity verification result of the visitor, determine the visitor's personnel-side permissions for the target resource object;
[0009] The device-side command permissions corresponding to the target resource object are determined based on the target fire impact address, and the personnel-side permissions are adjusted when both the personnel-side permissions and the device-side command permissions meet the control conditions corresponding to the access request.
[0010] Converting the original physical address of the target resource object to the target fire impact address includes:
[0011] Based on the original physical address of the target resource object, determine the physical ownership unit of the target resource object in the industrial site;
[0012] Based on the influence relationship between the current fire propagation front and the physical attribution unit, determine the fire influence unit currently corresponding to the target resource object;
[0013] The target fire-affected address is obtained by re-addressing the original physical address of the target resource object according to the fire-affected unit.
[0014] Based on the influence relationship between the current fire propagation front and the physical attribution unit, the fire influence unit currently corresponding to the target resource object is determined, including:
[0015] When the target resource object is located within the coverage area of the current fire propagation front, the directly affected unit corresponding to the current fire propagation front is determined as the fire impact unit;
[0016] When the target resource object is located outside the coverage area of the current fire spread front, and the target resource object is on at least one of the paths of smoke diffusion path, heat radiation influence path, evacuation path or fire linkage path, the associated influence unit corresponding to the path is determined as the fire influence unit.
[0017] The target fire impact address includes at least one of the following: fire event identifier, fire impact unit identifier, object role identifier, address version information, and valid time information;
[0018] The object role identifier is used to characterize the target resource object as belonging to one of the following categories under the current fire spread relationship: fire awareness object, evacuation impact object, smoke exhaust linkage object, hazard source associated object, fire water pressure guarantee object, or cross-regional control object.
[0019] Based on the current identity verification result of the visitor, determine the visitor's user-side permissions for the target resource object, including:
[0020] The set of authorized fire scene addresses corresponding to the visitor is determined based on the current identity verification result.
[0021] Determine whether the target fire-affected address falls within the set of authorized fire-affected addresses for personnel;
[0022] If the target fire-affected address falls within the set of authorized fire-affected addresses, then the access personnel's personnel-side permissions to the target resource object are determined based on the request action corresponding to the access request.
[0023] The set of authorized fire scene addresses for personnel is determined based on the emergency task information, task area, task validity time, and identity verification status of the personnel visiting the site.
[0024] The identity verification status is used at least to characterize whether the accessing person meets the continuous identity verification requirements under the zero-trust architecture when the access request occurs.
[0025] The method further includes:
[0026] If the target fire impact address does not fall into the personnel fire authorization address set, then the boundary association relationship between the fire impact unit corresponding to the target fire impact address and the fire impact unit corresponding to the personnel fire authorization address set is determined.
[0027] Based on the boundary association, the user-side permissions of the accessing personnel to the target resource object are adjusted, wherein the adjustment includes at least one of adjusting the control permission to the suggestion permission, adjusting the suggestion permission to the viewing permission, revoking the viewing permission, or rejecting the access request.
[0028] Determine the device-side command permissions corresponding to the target resource object based on the target fire-affected address, including:
[0029] When the target resource object is a controllable device object, the control sources, control actions, and address versions that the controllable device object is allowed to accept, execute, and use under the current fire propagation relationship are determined based on the target fire impact address.
[0030] The device-side command permissions are generated based on the control source, control action, and address version.
[0031] When both the personnel-side permissions and the device-side command permissions meet the control conditions corresponding to the access request, the personnel-side permissions are adjusted, including:
[0032] Determine whether the user's terminal permissions allow the user to initiate the control action corresponding to the access request;
[0033] Determine whether the device-side command permission allows the target resource object to accept the control source, control action, and address version corresponding to the access request;
[0034] If all are allowed, the user's access permissions to the target resource object are adjusted to the executable permissions corresponding to the control action;
[0035] If any of these permissions are not allowed, the user's access permissions to the target resource object will be adjusted to restricted permissions or denied permissions.
[0036] The method further includes:
[0037] When the current fire propagation front changes, the target fire impact address corresponding to the target resource object is re-determined;
[0038] If the redefined target fire impact address is inconsistent with the original target fire impact address, the personnel terminal permissions and the device terminal command receiving permissions shall be updated based on the redefined target fire impact address.
[0039] Specifically, when the target fire-affected address is within the address change stability window, the access personnel retain their viewing permissions for the target resource object, while restricting their control permissions over the target resource object.
[0040] Compared with the prior art, the beneficial effects of the present invention are:
[0041] This invention converts the original physical address of the target resource object into the target fire-affected address based on the current fire propagation front, changing the basis for permission judgment from static physical attribution to dynamic risk attribution. This allows the invention to reflect the true safety implications of resource objects in a fire scenario in real time, solving the problem of permission object mismatch caused by fire boundary migration and improving the accuracy of permission control. Attached Figure Description
[0042] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0043] Figure 1 This is a schematic diagram of the system architecture according to an embodiment of the present invention;
[0044] Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention;
[0045] Figure 3 This is a schematic diagram of resource object state transition according to an embodiment of the present invention.
[0046] Explanation of reference numerals in the attached figures:
[0047] 10. Central Server; 11. Data Acquisition Module; 12. Impact Analysis Module; 13. Address Encoding Module; 14. Permission Decision Engine; 20. Edge Computing Gateway; 30. Sensors; 40. Controllable Devices; 50. Operating Terminals; S100. Respond to Access Requests; S110. Convert Fire Scene Impact Addresses; S120. Determine Personnel Permissions; S130. Determine Device Permissions; S140. Dual-End Permission Verification; S150. Adjust / Execute Permissions; S160. Deny / Downgrade Permissions. Detailed Implementation
[0048] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0049] The embodiments mentioned herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of phrases in various places in the specification does not necessarily refer to the same embodiment, nor are they independent or alternative embodiments mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0050] Before providing a further detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.
[0051] Zero Trust Architecture: This refers to a security model whose core idea is not to trust any person, device, or system inside or outside the network boundary by default, but to determine each access request based on continuous authentication and authorization. In the specific scenario of this application, the architecture emphasizes that during industrial fire emergency response, every resource access and device control request must undergo dynamic, real-time context verification, rather than relying on one-time login authentication or static role assignment.
[0052] Resource objects refer to entities that can be accessed, invoked, or controlled within the industrial fire emergency platform. Specifically, they can be divided into three categories: data resources, such as real-time data streams collected by various sensors, including temperature, smoke concentration, and video images; equipment resources, such as remotely controllable equipment like smoke exhaust fans, sprinkler valves, fireproof roller shutters, and power distribution switches; and spatial resources, such as logical spaces defined in the 3D model, including fire compartments, evacuation routes, and hazardous source areas.
[0053] Fire impact address: This refers to a temporary access identifier dynamically assigned to a resource object under a specific fire propagation relationship. It differs from the fixed original physical address of the resource object that represents its physical installation location. Its core function is to characterize the object's actual safety attribution and risk association under the current fire situation. For example, a fan's physical address might be workshop A, but its fire impact address could be evacuation support unit B; the latter will be used as the basis for access control.
[0054] Personnel-side permissions refer to the operational authorization granted to an access user for a specific target fire-affected address after continuous identity verification through a zero-trust architecture. This authorization is dynamic and granular, and may include viewing permissions (only able to check device status), permission to suggest operations but not directly execute them, or control permissions to issue actual control commands.
[0055] Device-side command permissions refer to a set of boundary rules that a controllable device or its agent, such as an edge computing gateway, actively sets based on its assigned fire-affected address. These rules define which authorized entities, personnel, or systems the device is allowed to receive commands from, which specific types of actions it is permitted to perform, and which address version the commands must conform to. This transforms the device from a passive command receiver into a proactive security policy enforcer.
[0056] Please see Figure 2 This application aims to provide a dynamic identity and permission adaptive control method under a zero-trust architecture to address the fundamental problem that traditional permission control models based on static roles and fixed areas cannot adapt to the real-time changes in the security ownership of resource objects in dynamic and high-risk scenarios such as industrial fires. Existing technologies typically assume that the ownership of a device or data is fixed. However, in a fire, a fan located in zone A may become a critical piece of equipment affecting the evacuation of zone B due to smoke flow direction. Continuing to use the permission policy for zone A would pose a significant risk. The core of the method proposed in this application is that instead of directly granting permissions to people to access fixed objects, it uses a dynamic fire-affected address as an intermediary to dynamically bind permissions to the real-time fire situation.
[0057] In a basic implementation, the method responds to an access request initiated by an operator via terminal 50 for a specific target resource object, such as a controllable device. First, the method's initial step is to convert the original physical address of the target resource object into a target fire-affected address. The original physical address is static, such as the No. 2 exhaust fan in Workshop A, while the target fire-affected address is dynamic, requiring determination based on real-time information from the current fire propagation front. Essentially, this conversion decouples resource ownership from its physical location and re-anchors it to its functional role within the current fire propagation context. This approach addresses the technical challenge of changing resource security implications due to fire variations without updating the access control policy.
[0058] Furthermore, the method determines the user's access permissions to the resource objects represented by the aforementioned target fire-affected address based on the user's current identity verification result. Zero-trust architecture requires continuous identity verification, not just a single login action. Therefore, the system evaluates various factors such as the user's real-time identity status, task relevance, and location to determine their eligibility to operate on the resource under the specific fire situation. This ensures that the access subject is trustworthy at the moment of the request.
[0059] Meanwhile, a core principle of this method is that it considers not only human permissions but also object permissions. The method assigns a corresponding device-side command permission to the target resource object based on the target fire's impact location. This solves the problem in traditional solutions where anyone with permission can arbitrarily control equipment, ignoring the environmental risks inherent to the equipment itself.
[0060] It should be noted that the target resource object can be understood as a controllable device object or a target resource object with a command interface.
[0061] Finally, the method performs a two-way permission verification. Only when both the user's permissions and the target resource's device permissions meet the control conditions corresponding to the access request will the system ultimately approve the operation, adjust the user's permissions to an executable state, and issue the command. For example, if the user's permissions allow control and the device's permissions allow receiving control commands from the user, the operation is permitted only if both match. If either condition is not met—for example, if the user has permissions but the device refuses operation because it is on a critical evacuation route—the request will be rejected or downgraded. This two-way verification authorization mechanism greatly improves the security and accuracy of control commands in complex emergency scenarios.
[0062] Furthermore, in an optional implementation, the step of converting the original physical address of the target resource object into the target fire impact address is specified. The first step requires determining the physical affiliation unit of the target resource object within the industrial site based on its original physical address. The physical affiliation unit is static, such as the production workshop, floor, or pipeline system where a piece of equipment is located. This is the starting point for access control analysis and the identification of the resource's static attributes.
[0063] Subsequently, the system will determine the fire-affected unit corresponding to the target resource object based on the influence relationship between the current fire propagation front and the physical unit. The fire propagation front is dynamic and may be composed of various factors such as flames, smoke, and thermal radiation. Influence relationship analysis is the core intelligence of this solution; it determines whether and how a static physical unit is affected by the dynamic fire front. For example, if physical unit A itself is not on fire, but smoke from the fire front is drifting towards it through the ventilation duct, then unit A has a downstream smoke-related influence relationship with the fire front.
[0064] Finally, based on the determined fire impact units, the original physical addresses of the target resource objects are re-addressed to obtain the target fire impact addresses. This re-addressing process is equivalent to attaching a temporary label to the resource object reflecting its current fire impact role. Through these three steps—determining the physical attribution unit, determining the fire impact unit, and re-addressing—this method concretizes an abstract concept of address transformation into a clear and executable technical process, thereby solving the technical problem of how to accurately and reliably complete dynamic address mapping and achieving a fundamental shift in the basis for permission determination from static physical location to dynamic risk attribution.
[0065] Specifically, the physical attribution unit is used to characterize the static attribution of a target resource object under normal production conditions. This static attribution is typically determined by the building zoning, equipment ledger, pipeline number, control loop, sensor location, and data interface registration information of the industrial site. For data-type resources such as video cameras, temperature sensors, smoke sensors, and combustible gas sensors, their original physical address can include the factory building, floor, grid unit, installation point, and data acquisition direction. For controllable equipment resources such as smoke exhaust fans, fireproof roller shutters, sprinkler valves, fire pumps, access control systems, and power distribution switches, their original physical address can include the production workshop, fire compartment, access control cabinet, equipment loop number, linkage branch number, and control interface number. Since zero-trust access control requires quickly determining the current security attribution of the accessed object when an access request occurs, if the static physical root node of the target resource object is not determined first, the subsequent fire impact mapping will lack a positioning benchmark, easily leading to misidentification of devices with the same name, adjacent locations, or similar sensor data as the same resource object. Identifying the physical ownership unit is equivalent to establishing a traceable static anchor point for the target resource object, so that it can still maintain a one-to-one correspondence with the field entity, data interface and control loop when dynamic readdressing occurs in the future.
[0066] In this embodiment, an industrial site resource address table and a spatial partition index table can be pre-established. The industrial site resource address table records at least the resource object identifier, resource object type, original physical address, building unit, fire compartment, equipment system, control interface, data interface, and adjacent resource objects. The spatial partition index table records at least the spatial inclusion and connectivity relationships between production workshops, storage areas, pipe corridors, floors, evacuation routes, fire compartments, hazardous source areas, and fire protection facility coverage areas. When an access request carries a target resource object identifier, the original physical address of the target resource object is first read from the industrial site resource address table, and then its installation point, equipment system, and area affiliation are resolved from bottom to top according to the address hierarchy to determine the physical affiliation unit. For example, the original physical address "Plant A, Second Floor—Reaction Zone—Smoke Exhaust Branch 2—Fan F2" can be resolved as: Installation point: End of Smoke Exhaust Branch 2; Equipment system: Smoke Exhaust System; Spatial area: Reaction Zone; Fire compartment: Fire compartment: Second Floor, Plant A. If the same resource object has both spatial and system affiliations—for example, a fire pump belongs to a fire water system and affects multiple sprinkler branches—then both spatial and system affiliations are considered as candidate physical affiliation units, and the priority affiliation is determined based on the request action in the access request. When the request action is to view the operating status, the equipment installation area can be prioritized as the physical affiliation unit; when the request action is to start / stop control or branch switching, the linkage system or control loop to which it belongs can be prioritized as the physical affiliation unit. This approach avoids the problem of determining permissions solely based on the equipment installation location while ignoring the equipment's system influence scope.
[0067] Furthermore, the determination of physical attribution units can also incorporate address consistency verification to improve the reliability of resource identification before readdressing. Address consistency verification can be based on cross-validation of resource registration information, the location of the most recent heartbeat report, the access gateway number, and the on-site spatial coordinates. For example, if a device is recorded as belonging to workshop A in the logbook, but its access gateway number has long belonged to the fire control cabinet in workshop B, and its location coordinates reported in the last five minutes fall within the boundary of workshop B, then the resource object can be marked as having an address pending confirmation. In the permission judgment, instead of directly using a single physical attribution, both the logbook attribution and the access attribution are retained as candidate physical attribution units. For coordinate-type locations, a spatial tolerance threshold can be set to determine whether the coordinates fall within the boundary of a certain zone. This spatial tolerance threshold can be determined based on the modeling accuracy of the factory's electronic map. For example, when the point marking accuracy of the factory's 3D model is one meter, the spatial tolerance threshold can be two meters to cover sensor installation errors and drawing conversion errors. For gateway access devices, a stable access time threshold can be set. For example, if three consecutive heartbeat reports all come from the same control gateway, and the coverage time of the three heartbeats is no less than 30 seconds, then the control gateway is considered a valid access point. Through the above verification, the physical attribution identification deviation caused by ledger delays, equipment replacement, temporary access, or location entry errors can be reduced, so that the determination of subsequent fire-affected units is based on more reliable static resource anchor points.
[0068] Specifically, after determining the physical attribution unit, it is necessary to determine the current fire impact unit corresponding to the target resource object based on the influence relationship between the current fire propagation front and the physical attribution unit. The fire impact unit is not equivalent to the ignition area or the administrative area, but is determined by the actual safety implications of the current fire propagation front on the resource object. The fire propagation front can be composed of the flame front, smoke front, thermal radiation front, combustible gas front, toxic gas front, evacuation obstruction front, and fire linkage front. For the same target resource object, its physical location may not be within the flame area, but if it is located downstream of the smoke, in the adjacent thermal radiation zone, at a key node of the evacuation route, or within the influence range of a fire water branch, the target resource object should still be mapped to the corresponding fire impact unit. Using the fire impact unit, rather than the ignition point or a fixed area, as the basis for permission attribution ensures that permission judgment is consistent with the actual fire propagation relationship, preventing access personnel from obtaining resource permissions with cross-regional influence based solely on their original area identity.
[0069] In this embodiment, the current fire propagation front can be generated in the fire impact topology based on real-time fire status information. The fire impact topology can be composed of a spatial connectivity diagram, a ventilation and smoke extraction diagram, a fire water branch diagram, a power circuit diagram, an access control diagram, and a hazard source adjacency diagram. The spatial connectivity diagram is used to represent the spatial connectivity between workshops, passages, doorways, stairwells, pipe corridors, and equipment rooms; the ventilation and smoke extraction diagram is used to represent the airflow relationship between air ducts, smoke extraction branches, fans, smoke extraction outlets, and make-up air outlets; the fire water branch diagram is used to represent the water supply influence between fire pumps, main pipes, branch pipes, sprinkler valves, and terminal sprinklers; the power circuit diagram is used to represent the power supply influence between power distribution switches, emergency lighting, sensors, and control cabinets; and the hazard source adjacency diagram is used to represent the adjacency relationship between hazardous chemical storage locations, flammable material storage areas, high-temperature equipment, and pressure vessels relative to the fire front. Real-time fire status information can include temperature values, smoke concentration, combustible gas concentration, thermal imaging anomaly areas, video flame recognition areas, smoke extraction equipment operating status, sprinkler operation status, personnel positioning, and access control opening and closing status. By projecting the aforementioned real-time data onto the fire impact topology, different types of propagation fronts can be obtained. For example, if the smoke concentration in two consecutive adjacent spatial units exceeds twice their respective background values, and the peak smoke concentration occurs later along the channel direction, this direction can be identified as the smoke propagation direction. If the temperature outside a hazardous area rises by more than 15 degrees Celsius within one minute, and the abnormal thermal imaging area approaches the boundary of the hazardous area, the outer edge of the hazardous area can be identified as the thermal radiation impact front. If the smoke concentration in an evacuation route reaches the preset evacuation restriction concentration, and the access control or fireproof roller shutter status increases the detour distance for personnel, this route can be identified as the evacuation impact front. The above thresholds can be determined based on the historical background values of on-site sensors and fire safety management requirements. For example, the smoke concentration can be the average value of the most recent 24-hour normal production period as the background value, the temperature change threshold can be determined by adding a reserve amount to the historical temperature fluctuation limit of the same area, and the evacuation restriction concentration can be calculated by converting the set values of evacuation visibility or smoke alarm level in the enterprise's fire plan.
[0070] Furthermore, the influence relationship between the current fire propagation front and the physical attribution unit can be identified according to four categories: direct coverage, downstream path, reverse action influence, and decision support. Direct coverage refers to a physical attribution unit located within the coverage area of the fire, smoke, heat radiation, toxic gases, or combustible gases front; in this case, the target resource object is usually mapped to a directly affected unit. Downstream path relationship refers to a physical attribution unit not within the front coverage area but located downstream of a smoke diffusion path, ventilation path, smoke exhaust path, personnel evacuation path, or fire water pressure transmission path; in this case, the target resource object is mapped to the corresponding downstream affected unit. Reverse action influence relationship refers to a target resource object's own start-stop, shutdown, power outage, smoke exhaust, sprinkler, or access control actions that reversely change the status of the fire propagation front or other areas; in this case, the target resource object should be mapped to a linked influence unit, rather than simply retaining its original physical attribution. Decision support relationship refers to a target resource object that, although not covered by the fire front and whose actions do not directly change the front, has data necessary for determining the status of hazard sources, evacuation paths, or adjacent areas; in this case, it can be mapped to a decision support influence unit. For example, if a smoke exhaust fan is installed in area A, but its smoke exhaust direction affects the evacuation route in area B, then the smoke exhaust fan belongs to the reverse action influence relationship and can be mapped to the evacuation influence unit in area B. Similarly, if a temperature sensor is installed outside a hazardous chemical storage facility, and the flame has not yet entered the facility, but the thermal radiation front is already approaching the location, then the temperature sensor belongs to the decision support relationship and can be mapped to the hazard source front judgment unit. To confirm influence relationships, duration conditions can be set to avoid mismapping caused by instantaneous disturbances. For example, the smoke front influence relationship requires that the concentration increase condition be met for two consecutive sampling periods, with each sampling period lasting ten to thirty seconds. The thermal radiation influence relationship requires that the thermal anomaly area expands towards the target physical unit in three consecutive thermal imaging frames. The linkage influence relationship can be confirmed without relying on instantaneous sensor values, but based on whether there is a registered influence link between the control action and the downstream area in the equipment linkage topology. By classifying and identifying influence relationships, target resource objects can be mapped to fire scene influence units that match their current fire scene effect, rather than mechanically using the original area permissions.
[0071] Specifically, after identifying the fire-affected unit, the original physical address of the target resource object needs to be re-addressed based on that fire-affected unit to obtain the target fire-affected address. Re-addressing does not overwrite or delete the original physical address; instead, it generates a temporary access address for the target resource object that corresponds to the current fire propagation relationship, while retaining the original physical address as a resource entity index. This temporary access address is used for subsequent personnel access control and device command access control, transforming access requests from accessing a specific device or data in a certain area to accessing a specific role object under a fire-affected unit. For example, the original physical address of a fan may still be the second fan (F2) in the smoke exhaust branch of Workshop A, but when its operation affects the evacuation status of Channel B, its target fire-affected address can be re-addressed to the smoke exhaust control object in the evacuation affected unit of Channel B under fire event E1. In this way, personnel access control is no longer solely based on equipment management permissions in Workshop A, and device command access control is no longer solely based on the original control list, but rather both are adapted around the target fire-affected address.
[0072] In this embodiment, the target fire impact address can be composed of a fire event identifier, a fire impact unit identifier, an object role identifier, an original resource index, address version information, and valid time information. The fire event identifier distinguishes different fire events, preventing address confusion for the same resource object in different fire situations. The fire impact unit identifier indicates the fire impact boundary currently attached to the target resource object, such as a downstream smoke impact unit, a thermal radiation front unit, an evacuation maintenance unit, a hazard source isolation unit, a fire water pressure guarantee unit, or a smoke exhaust linkage unit. The object role identifier indicates the role of the target resource object within the fire impact unit, such as a fire awareness object, a smoke exhaust control object, an evacuation maintenance object, a hazard source association object, a linkage blocking object, or a decision support object. The original resource index refers back to the target resource object's original physical address and equipment ledger number, ensuring that the actual equipment or data interface can still be located after the permission address changes. The address version information records the re-addressing order of the target resource object in the same fire event. The valid time information limits the effective scope of the target fire impact address. For example, in fire event E1, the target fire impact address of smoke exhaust fan F2 can include fields such as E1, B-channel evacuation impact unit, smoke exhaust control object, F2 original resource index, version three, and validity period of ninety seconds. When the smoke direction changes again and causes the fan to affect the hazard source isolation of storage area C, the address version is incremented, the fire impact unit and object roles are updated accordingly, and the original resource index remains unchanged. Setting the address version allows subsequent control requests to identify whether they are generated based on expired addresses, preventing access personnel or equipment from using old fire impact relationships to execute control actions when accessing the gateway.
[0073] Furthermore, the readdressing results can be written to the fire impact address mapping table and kept synchronized with the permission policy engine and device access gateway. The fire impact address mapping table records at least the original physical address, the target fire impact address, the mapping generation time, the mapping basis, the address version, and the mapping validity status. The mapping basis can record the type of impact relationship that triggered the readdressing, such as direct coverage, downstream path, reverse action impact, or decision support; it can also record the data sources involved in the judgment, such as smoke sensors, thermal imaging equipment, fan status, access control status, or fire water pressure data. To avoid frequent address switching caused by short-term fluctuations in the fire propagation front, an address stability window can be set. The address stability window can be determined based on the fire data sampling cycle and the response time of on-site equipment. For example, when the fire data is updated every ten seconds and the smoke exhaust fan response takes an average of twenty seconds, the address stability window can be set to thirty to sixty seconds. Within this stability window, if the newly identified fire impact unit only switches between adjacent units with the original fire impact unit, the viewing permission under the original target fire impact address can be temporarily not released, and the control permission upgrade can be restricted until the new fire impact unit is confirmed to be stably present after several consecutive sampling cycles. For high-risk objects, such as hazardous chemical storage locations, main fire water equipment, main evacuation passage access control, and cross-area smoke exhaust equipment, stricter stability conditions can be adopted, such as requiring that the control address switch is completed only after three consecutive sampling cycles point to the same fire impact unit.
[0074] In another alternative implementation, the logic for determining fire-affected units based on influence relationships is refined. This application mainly divides it into two cases. The first is direct influence: when the target resource object itself is located within the coverage area of the current fire propagation front, for example, a temperature sensor is located in the flame zone, then the directly affected unit defined by this fire propagation front, such as the combustion core unit in zone A, is determined as the fire-affected unit of the resource object. In this case, the role and risk of the resource object are clear and direct.
[0075] The second type is related impact, which is more complex and common: when the target resource is located outside the fire propagation front coverage area, but its physical location or function affects certain critical paths, it will also be included in the control. These paths include at least: smoke diffusion paths, such as exhaust fans affecting downstream passages; heat radiation impact paths, such as high-temperature areas threatening adjacent hazardous chemical tanks; personnel evacuation paths, such as an access control point located on the only escape route; or fire linkage paths, such as the opening and closing of a valve affecting the fire water pressure in another area. In this case, the system will identify the related impact units corresponding to these paths, such as the smoke impact unit of passage B and the heat radiation early warning unit of the hazardous chemical storage tank area, as the fire impact units of the resource.
[0076] In one optional implementation, the data structure for the target fire impact address is specifically defined. The target fire impact address can be a structured string or data object containing multiple fields, including at least: a fire event identifier to distinguish different fire events, such as 20231026-WH01; a fire impact unit identifier, such as B-Corridor-Evac-Unit; an object role identifier; address version information; and valid time information.
[0077] The object role identifier is used to characterize the specific role played by the target resource object in the current fire propagation relationship. For example, it can be identified as: fire perception object, such as temperature sensor in the fire scene; evacuation impact object, such as access control or fan on evacuation routes; smoke exhaust linkage object; hazard source associated object, such as hazardous chemical storage tank near the fire scene; fire water pressure protection object, such as valve on critical fire pipelines; or cross-area control object, etc.
[0078] Furthermore, in an optional implementation, the internal process of determining the access permissions S120 of the accessing personnel to the target resource object based on the current identity verification result of the accessing personnel is described. The process first requires determining a set of authorized fire scene addresses for the accessing personnel based on their current identity verification result. This set is a list of all fire scene impact units that the personnel are authorized to operate on at the current moment, based on their identity and task.
[0079] Subsequently, the system performs a crucial matching check: determining whether the target fire impact address of the target resource object falls within the personnel's authorized fire impact address set. This is essentially a matching process; the personnel's authorized address set is like a keychain, while the target's fire impact address is like a lock.
[0080] If a match is successful, meaning the target fire-affected address falls within the authorized address set of personnel, the system will further determine the specific personnel permissions for the target resource object based on the specific actions of this access request, such as viewing, opening, or closing, combined with preset policies. For example, viewing and opening may be allowed, but closing may be prohibited. This method standardizes and streamlines the process of determining personnel permissions, solving the problems of ambiguous permission determination logic and reliance on manual judgment, and achieving automation and precision in personnel permission determination.
[0081] Understandably, the specific matching process can be determined by referring to the address field matching results between the target fire impact address and the personnel fire authorized address set, the subordinate relationship of the impact unit, the boundary association relationship, and the address validity time relationship. This ensures that personnel permissions are no longer simply given by personnel positions or static area permissions, but are jointly limited by the fire impact boundary that the accessing personnel can currently handle and the fire impact boundary to which the target resource object currently belongs.
[0082] Specifically, the personnel fire scene authorization address set does not directly read the fixed area permissions from the pre-defined roles of the accessing personnel. Instead, based on the current identity verification result at the time of the access request, it converts the accessing personnel's identity status, task status, and fire scene handling boundary into several fire scene authorization addresses that can be recognized by the permission policy. The current identity verification result can include at least one of the following: personnel identity authentication status, terminal trust status, task work order status, emergency response role, personnel real-time location, task validity time, command link source, and the most recent continuous verification result. Since the permission objects in industrial fire scenarios have been converted from the original physical address to the fire scene impact address, if fixed identity fields such as workshop director, environmental protection personnel, and equipment maintenance personnel are still used as the direct authorization basis, it is easy to encounter situations where the personnel's identity is valid, but their task boundary is inconsistent with the current fire scene impact boundary of the target resource object. Therefore, before determining the personnel's permissions, it is necessary to first convert the accessing personnel's identity verification result into a personnel fire scene authorization address set at the same semantic level as the target fire scene impact address, so that both the personnel authorization object and the resource object are within the fire scene impact address space for comparison.
[0083] In this embodiment, the set of authorized personnel fire scene addresses can be generated according to the structure of fire scene event identifier—authorized impact unit—authorized action range—effective time window—address version range. Specifically, the current emergency task information is first read based on the access personnel identifier carried in the access request, and the validity of the emergency task is determined based on the current identity verification result. Emergency task information may include task number, task type, task area, target, task start time, expected duration, task source, and authorized personnel. If the access personnel's identity authentication is successful, the terminal access status is valid, the task work order has not timed out, and the spatial consistency requirement between the personnel location and the task area is met, one or more authorized personnel fire scene addresses are generated based on the fire scene impact unit corresponding to the task area under the current fire propagation relationship. The spatial consistency requirement can be determined based on the plant's positioning accuracy and task type. For example, when the indoor Bluetooth or UWB positioning error is one to three meters, the allowable deviation between the personnel location and the boundary of the task area can be set to five meters. For remote command personnel, it is not required that their location be within the on-site area, but rather that their command link source, task number, and remote terminal trust status all meet the verification conditions. The effective time window can be determined by the task start time and the expected duration of the task, or it can be dynamically shortened according to the stage of the fire event. For example, the effective time window for the address of on-site verification tasks can be five to fifteen minutes, and the effective time window for the address of equipment control tasks can be thirty seconds to three minutes, and it should be re-verified every time a control request occurs.
[0084] Furthermore, the set of authorized addresses for personnel in a fire scene can generate addresses of different granularities based on the different responsibilities of the personnel in the fire scene response chain. For on-site response personnel, the authorized address can be refined to a specific fire scene impact unit, such as the fire scene event E1—B channel evacuation impact unit—view and suggested actions—valid for sixty seconds; for equipment maintenance personnel, the authorized address can be refined to a specific equipment linkage impact unit, such as the fire scene event E1—smoke exhaust linkage unit—status view and restricted control actions—valid for ninety seconds; for command personnel, the authorized address can cover multiple adjacent fire scene impact units, but control actions can still be restricted by the unit role. If the current identity verification result shows that the visitor's identity is valid but the terminal's trust status is reduced, such as the terminal location remaining unchanged for a long time, the terminal network jumping to an unregistered gateway, or the continuous verification interval exceeding the preset time, then the view-type authorized address can be retained, and the control-type authorized address will not be generated temporarily. The continuous verification interval can be determined based on the access risk. For ordinary viewing requests, it can be set to three to five minutes, and for control requests, it can be set to thirty to sixty seconds. If no new identity verification result is obtained within this interval, the corresponding personnel's fire scene authorization address will be marked as pending refresh.
[0085] Specifically, after obtaining the set of authorized firefield addresses for personnel, it is necessary to determine whether the target firefield impact address of the target resource object falls within this set. This determination should not be based solely on identical strings, as firefield impact addresses may involve various relationships such as parent-child impact units, adjacent impact units, linked impact units, and decision support units. Whether a target firefield impact address falls within the set of authorized firefield addresses should be determined comprehensively by considering the address field, firefield impact unit hierarchy, boundary association type, and effective time. This matching method ensures that resource objects within the same firefield impact boundary are quickly authorized, while also preserving limited processing space when the target resource object is on adjacent, downstream, or linked boundaries. This avoids rejecting all incompletely identical addresses, preventing on-site personnel from obtaining necessary firefield response data.
[0086] In this embodiment, a hierarchical matching method can be used to determine the relationship between the target fire-affected address and the set of authorized personnel fire-affected addresses. The first layer is event consistency matching, meaning that the fire event identifier in the target fire-affected address should be consistent with the fire event identifier in the authorized personnel fire-affected address, preventing access personnel from using permissions under other fire-related tasks to access the current event resource. The second layer is impact unit matching, which determines whether the fire-affected unit identifier in the target fire-affected address is the same as the authorized impact unit identifier in the set of authorized personnel fire-affected addresses, or whether it belongs to a subordinate impact unit of that authorized impact unit. For example, when personnel are authorized to handle the B-channel evacuation impact unit, the smoke monitoring object, access control object, and smoke exhaust status object under that unit can be matched; if the target resource object is re-addressed as the B-channel evacuation impact unit—smoke exhaust control object, it can be determined that it falls into that authorized impact unit. The third layer is object role matching, which determines whether the target resource object's object role falls within the scope of object roles that the accessing personnel are allowed to access under the corresponding authorized address. For example, on-site evacuation personnel can access evacuation maintenance objects, smoke monitoring objects, and access control objects, but do not automatically gain control over hazard isolation objects or fire water pressure protection objects. The fourth layer is valid time matching, which means that the valid time of the target fire impact address and the valid time of the personnel fire authorization address should overlap, and the access request should occur within the overlapping time. If the target fire impact address has been updated to a new version, but the personnel fire authorization address still corresponds to the old version, the match is not directly considered successful, but rather enters the address version review process.
[0087] Furthermore, for target fire impact addresses that do not fully fall within the personnel fire scene authorization address set but have a boundary association with them, association matching processing can be set up. Association matching is applicable when the target resource object does not belong to the current handling unit of the visiting personnel, but there is a necessary fire propagation link between it and the visiting personnel's handling task. For example, if the target resource object is located upstream of the smoke, downstream of the smoke exhaust, in the evacuation adjacency zone, or within the hazard warning boundary of the authorized impact unit, the visiting personnel may need to view the status of the target resource object to complete the current task. Association matching can be processed differently according to the boundary association type: when the target fire scene impact address and the personnel authorization address are in the same impact unit or have a parent-child impact unit relationship, access to the action permission judgment can be allowed; when the two are adjacent evacuation units, upstream and downstream smoke units, or smoke exhaust linkage units, only viewing or suggestion permissions can be allowed; when the two only have a weak spatial adjacency relationship and there is no current fire propagation link to support them, access can be denied. Boundary relationships can be obtained from the fire impact topology and confirmed in conjunction with the propagation direction. For example, upstream units of smoke have a high correlation with downstream evacuation decisions, while adjacent units with opposite smoke propagation directions should not be automatically granted permissions. To avoid misjudgments in association matching caused by instantaneous boundary fluctuations, boundary relationships can be required to remain consistent over two consecutive sampling periods, with each sampling period lasting ten to thirty seconds. For control requests, the version of the target fire impact address corresponding to the association can also be required to be consistent with the version in the latest mapping table. Through the combination of hierarchical matching and association matching, the matching between the target fire impact address and the set of authorized personnel fire addresses has clear rules and can adapt to temporary changes in resource object ownership during fire boundary migration.
[0088] Specifically, when the target fire-affected address falls within the set of authorized fire-affected addresses, or is identified through association matching as having an authoritative relationship with the current personnel handling boundary, it is necessary to determine the personnel-side permissions of the accessing personnel to the target resource object based on the request action corresponding to the access request. Personnel-side permissions should not be simply expressed as allow or deny, but should be refined based on the object role of the target resource object, the risk level of the request action, the scope of the accessing personnel's authorized actions, and the coordination relationship with the device's commanded permissions. Request actions can include viewing, downloading, calling, annotating, suggesting, opening, closing, resetting, switching, transferring, and confirming; among these, viewing, calling, and suggesting are usually information-related or decision support actions, while opening, closing, resetting, and switching are usually control-related actions. Since the object role of the same target resource object may change under different fire-affected addresses, the same accessing personnel may also have different permissions for the same device at different times. For example, a person may have opening permission for a smoke exhaust fan under the smoke exhaust unit in this area, only suggestion permission under an adjacent evacuation-affected unit, and only status viewing permission under a hazard source isolation unit.
[0089] In this embodiment, personnel permissions can be determined step-by-step based on the risk level of the action. For viewing actions, if the target fire-affected address and the set of authorized personnel fire-affected addresses satisfy event consistency, impact unit matching, or association matching, and the access request occurs within a valid time window, then the personnel permission can be determined as viewing permission. For suggestion actions, in addition to meeting the conditions for viewing actions, the emergency task type of the accessing personnel must also have a handling association with the object role of the target resource object. For example, evacuation personnel can submit handling suggestions for access control, smoke exhaust status, and passage smoke data, while equipment maintenance personnel can submit control suggestions for equipment start / stop status. For control actions, in addition to meeting the condition that the target fire-affected address falls within the set of authorized personnel fire-affected addresses, the current identity verification result must contain a valid control authorization status, and the requested action must fall within the authorized action range corresponding to the personnel fire-affected address. The control authorization status can be determined based on the control authorization mark in the task work order, the command link confirmation result, the terminal trust status, and the time of the most recent identity verification; for example, if the control authorization mark is valid, the command link source is a registered emergency command node, and the terminal passes continuous verification within the last sixty seconds, then the control authorization status can be considered satisfied. If the above conditions are not met, even if the access user's identity is successfully authenticated, control permissions will not be granted directly. Instead, the user's permissions will be set to suggested permissions or viewing permissions.
[0090] In another alternative implementation, the basis for generating the set of authorized fire scene addresses for personnel is further explained. This set is not generated out of thin air, but rather based on a comprehensive set of dynamic information closely related to the personnel. This information includes at least: the emergency task information currently assigned to the personnel, such as whether they are responsible for firefighting, evacuation, or equipment support; the clearly defined area of effect of the task; the effective time of the task; and the personnel's identity verification status.
[0091] Identity verification status is a core aspect of the zero-trust architecture. It characterizes whether, at the moment an access request is initiated, the user still meets the continuous identity verification requirements under this architecture. This may include multi-dimensional indicators such as biometric matching, device security, and any abnormal behavior. By tightly binding a user's permission set to their specific tasks, spatiotemporal scope, and real-time trust status, this method ensures that permission granting follows the principles of least privilege and on-demand authorization. It solves the problems of overly broad and easily abused permissions caused by traditional static role-based authorization, such as all firefighters having permission A. This makes permission allocation more dynamic, precise, and secure.
[0092] In an alternative implementation, processing logic is also considered when the target fire-affected address does not fall within the set of authorized personnel fire-affected addresses. In traditional access control models, this typically translates to a hard access denial. However, in complex emergency collaborative operations, indiscriminate denial can hinder necessary information sharing and cross-regional cooperation. Therefore, this method provides a more flexible processing mechanism.
[0093] When the aforementioned mismatch occurs, the system will not immediately reject the application. Instead, it will first determine whether there is a boundary association between the fire impact unit corresponding to the target fire impact address and the fire impact units covered by the personnel fire authorization address set. For example, Personnel A is responsible for the western fire zone, while the target equipment belongs to the northern fire zone, but the western and northern zones are geographically adjacent. This adjacency relationship is a boundary association.
[0094] Based on the identified boundary relationships, the system can dynamically adjust the access permissions of personnel instead of directly denying them. This adjustment can be multi-layered, for example: downgrading the originally requested control permission to a suggestion permission, allowing personnel to propose operational suggestions for review and execution by authorized personnel; or downgrading the control permission to a viewing permission, allowing them to view device status to understand the situation in the surrounding area; in certain high-risk situations, viewing permissions can also be revoked, or access can be ultimately denied. This method, while ensuring security, greatly enhances the flexibility and coordination of emergency response, solving the problems of information silos and collaboration barriers caused by strict regional divisions in traditional models.
[0095] Furthermore, in an optional implementation, the implementation method of determining the device-side command permission S130 corresponding to the target resource object based on the target fire impact address is described in detail. This step is crucial for achieving proactive device safety. When the target resource object is a controllable device object, such as a smoke exhaust fan, the system will generate a set of explicit command rules for it based on its current target fire impact address.
[0096] This set of rules defines at least three aspects: the sources of control that the equipment is allowed to receive under the current fire spread relationship, such as whether it comes from the command center or from the person in charge of a specific area; the control actions that are allowed to be performed, such as allowing opening and speed adjustment but prohibiting closing; and the address versions that are allowed to be used to prevent instructions based on outdated fire information from being executed.
[0097] Based on these three dimensions, the system generates the device's current command permissions. These permissions can be stored in the device itself or in the edge computing gateway 20 it is connected to. In this way, the method pushes the permission policy down to the device, transforming the device from a passive command executor into an intelligent agent capable of actively verifying the legitimacy of commands based on its own fire scene environment. This solves the fundamental deficiency of relying solely on cloud or platform-side permission control, which cannot prevent unauthorized commands from reaching the field devices.
[0098] In another optional implementation, the final dual-end permission collaborative verification and permission adjustment logic is described in detail. When a control request is initiated, the permission decision engine executes a strict double-check process. First, it determines whether the user's permissions allow the user to initiate the control action corresponding to the access request. For example, if a user requests to turn off a fan, but their user permissions are only for viewing, this check will fail.
[0099] Secondly, it determines whether the device's command permissions allow the target resource object to accept the control source, control action, and address version corresponding to this access request. For example, if a person requests to shut down a fan, the command source, action, and address version are all included in the request. The device or its gateway will check its own command permission table to see if the source, action, and address version are allowed, and if they match. If the device's current role is evacuation route protection, its command permissions may have already prohibited the shutdown action, and this check will also fail.
[0100] Only when both of the above checks pass will the system ultimately adjust the user's permissions on the target resource object to executable permissions corresponding to the control action, and allow the command to be issued. If either check fails, the user's permissions will be adjusted to restricted permissions, such as downgrading to suggested permissions or denying permissions. This closed-loop logic of dual verification and dynamic adjustment constitutes the core security mechanism of this method. It ensures that every control operation is performed by the right person, at the right time, on the right device, and with the right operation, thereby resolving various risks of misoperation that may be caused by unilateral authorization and improving the level of control security.
[0101] To address the rapidly changing nature of fire scenes, the method in this application embodiment also includes a complete dynamic update mechanism. When the system detects a change in the current fire propagation front through sensors or other devices, such as a change in fire spread or smoke direction, it triggers a reassessment of permissions. The system will then re-determine the corresponding target fire impact address for the affected target resource objects.
[0102] If the redefined target fire impact address differs from the previous address—for example, a fan changes from being used for fire suppression in Zone A to being used for smoke extraction in Zone B—the system will trigger a chain update: based on this new target fire impact address, the personnel's access permissions and the device's own command permissions will be automatically updated. This forms a dynamic, adaptive closed loop of situational change, address update, and dual-end permission refresh.
[0103] Furthermore, to prevent repeated granting and revocation of control permissions due to small and frequent fluctuations at the fire boundary, which could cause system instability or operational confusion, this method introduces the concept of an address change stabilization window. When the fire-affected address of a resource object switches back and forth between two or more states within a short period of time, such as a few seconds, the system can temporarily place the object's control permissions in a restricted state, disallowing high-risk operations, while retaining its viewing and suggestion permissions. The corresponding stable control permissions are then granted only after the address stabilizes in a certain state for more than a preset time window. This mechanism, by introducing control damping, solves the jitter problem of dynamic systems at boundary states, ensuring the real-time performance and stability of the entire control method.
[0104] Please see Figure 1 This application also provides a dynamic identity and permission adaptive control system under a zero-trust architecture, which is the physical carrier of the above-described method. The system may include a central server 10, at least one edge computing gateway 20, multiple sensors 30, multiple controllable devices 40, and at least one operating terminal 50.
[0105] The central server 10 is used to execute core permission decisions and deploys core software functional modules on it. Specifically, it may include: a data acquisition module 11, responsible for aggregating real-time fire data from various sensors 30 distributed throughout the industrial site, such as temperature, smoke detectors, infrared thermal imaging, and gas detectors, as well as third-party systems, such as personnel positioning systems; an impact analysis module 12, which has a built-in digital twin model or impact topology map of the industrial site, and can calculate the fire propagation front in real time based on the data transmitted from the data acquisition module 11, and analyze its impact on various resource objects; an addressing module 13, which dynamically generates or updates the fire impact address for each affected resource object based on the output of the impact analysis module 12; and a permission decision engine 14, which is the core of policy execution, responsible for receiving access requests from the operation terminal 50, and executing the entire process described above of determining personnel permissions, determining device-side command permissions, and conducting dual-terminal collaborative verification.
[0106] The edge computing gateway 20 is deployed close to the field controllable device 40, serving as a bridge and front-end node between the central server 10 and the field devices. It receives control commands and updated device-side permission policies issued by the permission decision engine 14, and is responsible for performing the final command verification and distribution locally. Deploying some verification logic to the edge reduces network latency and provides basic protection capabilities in extreme cases of communication interruption with the central server 10.
[0107] Sensor 30 is the system's sensing endpoint, providing all the raw data needed for system decision-making. Controllable device 40 is the system's execution endpoint, the final object of access control, such as smoke exhaust fans, fire pumps, and fireproof roller shutters. Operating terminal 50 is the human-machine interface, through which emergency responders initiate access requests and receive system feedback.
[0108] refer to Figure 3 , Figure 3 This is a schematic diagram of resource object state transitions provided in an embodiment of this application.
[0109] like Figure 3 As shown, resource objects are in a normal state when they are not affected by fire propagation. In this state, the resource objects are subject to regular access control rules based on their original physical addresses. This normal state does not mean that the resource objects are not registered or controlled, but rather that their access control is still determined by static addresses such as equipment ledgers, spatial partitions, data interfaces, or control loops. For example, smoke extraction equipment, access control equipment, or environmental sensors installed in Zone A can continue to be access-controlled as resource objects in Zone A when they are not affected by the fire propagation front.
[0110] When the current fire propagation front covers Area A, resource objects located in Area A or directly related to the fire propagation in Area A are transformed from their normal state to Fire Impact Unit A. In this state, resource objects are no longer subject to permission checks solely based on their original physical addresses, but rather on the target fire impact address corresponding to Fire Impact Unit A. Accordingly, whether an accessing person has original management permissions in Area A is not the sole criterion; it is also necessary to determine whether their current identity verification result can generate a personnel fire authorization address corresponding to Fire Impact Unit A. If the fire is extinguished, or the fire propagation front no longer covers Area A and the relevant impact relationship meets the conditions for extinguishment, the resource objects can be restored from Fire Impact Unit A to their normal state, and personnel permissions will be restored to regular permissions or their temporary emergency permissions will be revoked.
[0111] During fire control, resource objects may shift from fire-affected unit A to fire-affected unit B due to the migration of fire-affected boundaries. For example, although the original physical address of a resource object may still be located in area A, if the smoke path has already affected area B, or if the linkage between smoke extraction, isolation, ventilation, and sprinkler systems causes the control consequences of the resource object to shift to area B, then the target fire-affected address of the resource object needs to be updated from fire-affected unit A to fire-affected unit B. This conversion process does not involve a change in the physical location of the resource object, but rather a change in the ownership of permissions for the resource object under the current fire propagation relationship. Correspondingly, the viewing, suggestion, or control permissions originally granted based on fire-affected unit A need to be re-verified; if the personnel's authorized fire-affected address set does not cover fire-affected unit B, their personnel permissions can be downgraded, frozen, or revoked to prevent personnel from continuing to access or control target resource objects that already have the attributes of area B based on mismatched area A authorization.
[0112] Furthermore, when the transition between fire-affected unit A and fire-affected unit B is triggered by the smoke path affecting area B, the target resource object can be marked as a downstream associated object of the smoke. When the transition is triggered by a change in fire-affected area, the target resource object can be marked as a linked affected object or a decision support object in the new fire-affected unit. Through the above state transitions, the ownership of resource objects can be dynamically updated according to the current fire propagation front, smoke path, and impact resolution status. This ensures that personnel permissions are no longer fixed to the original physical address of the resource object, but are consistent with the actual fire-affected role currently assumed by the resource object.
[0113] In one specific implementation, combining the technical features defined in all the preceding claims yields a complete embodiment incorporating the aforementioned technical features. Imagine a large chemical plant's central control room deploying the control system described in this application. The system has pre-built a resource object library for the entire plant and an influence topology including spatial connectivity, pipeline routing, and ventilation relationships. One day, a reactor in workshop A experiences a small-scale fire due to overheating.
[0114] The system workflow is as follows: First, multiple sensors 30 located in Workshop A detect abnormal data in temperature, smoke, and combustible gas. Data acquisition module 11 captures and reports this data. Impact analysis module 12 immediately performs a topological analysis to determine the initial range of the fire propagation front and, based on the ventilation system drawings, predicts that the smoke will affect the adjacent B evacuation route along exhaust branch 2. Accordingly, addressing module 13 performs the following operations: re-addressing sensors, valves, and other objects in Workshop A as Fire Event E01 - Directly Affected Units in Area A - Fire Sensing / Controlled Objects - V1; simultaneously, re-addressing the critical exhaust fan F2, which is physically located in Workshop A but belongs to exhaust branch 2, as Fire Event E01 - B Route Evacuation Affected Units - Evacuation Affected Objects - V1. The address includes event, unit, role, and version information.
[0115] At this time, an operator in Workshop A, whose identity, task, initial response in Workshop A, and location have all passed zero-trust continuous verification, and whose authorized fire address set mainly covers the directly affected units of Area A, requests to shut down fan F2 through terminal 50, intending to cut off ventilation to stop the fire. The request reaches the authorization decision engine 14. The engine finds that the target fire impact address of fan F2 is the evacuation impact unit of passage B, which does not fall into the operator's authorized address set. However, the engine detects a boundary relationship between the smoke exhaust paths of Area A and passage B, so it does not directly reject the request, but automatically downgrades the operator's authorization to F2 from control to suggestion, and pushes a prompt to the operator: the operation may affect the evacuation of passage B, and has been changed to suggestion.
[0116] Meanwhile, the commander in charge of the B-channel evacuation, whose authorized fire scene address set includes the B-channel evacuation impact unit, requests the activation of fan F2 to ensure the channel remains smoke-free. The authorization decision engine 14 verifies the request and generates a control command containing the target fire scene impact address, ...-V1, which is then sent to the edge computing gateway 20 where fan F2 is located. Upon receiving the command, the edge computing gateway 20 performs device-side authorization verification: it checks if the current address version of F2 stored in its own database is V1, if the command source is an authorized evacuation command link, and if activation is an action permitted by the evacuation impact object. After all checks pass, the command is finally executed, and fan F2 starts.
[0117] Minutes later, the fire spread, and the fire front in Workshop A directly threatened the motor safety of fan F2. Impact analysis module 12 detected this change, and addressing module 13 immediately updated the address of fan F2 to Fire Event E01-Directly Affected Unit in Area A-Hazard Source Associated Object-V2. The address version changed from V1 to V2. The system automatically refreshed dual-end permissions. At this point, even if the evacuation commander in Channel B issued an activation command again, the command would still carry outdated V1 address information. When the edge computing gateway 20 performed device-side command permission verification, it would find that the address version V1 of the command did not match the current address version V2 of the device, thus refusing to execute the command and reporting an anomaly. Simultaneously, the system might transfer control of F2 to the fire fighting team, whose new task authorization set would include the updated address. During the brief moment of address switching, the address stabilization window mechanism activated, temporarily freezing control of F2, allowing only viewing to prevent confusion. Through this series of seamlessly connected dynamic processes, this embodiment ensures that the operation of each key piece of equipment is accurate, safe, and conforms to the globally optimal emergency strategy during the complex evolution of a fire scene. Compared with the static permission model, its overall technical effect has been significantly improved.
[0118] The technical solutions provided in this application can be widely applied to various industrial scenarios with complex spatial structures and high levels of automation, especially in industries with high risks such as fires, explosions, and toxic gas leaks. For example, in large-scale steel smelting enterprises, this method can be used to dynamically coordinate ventilation, power outages, and material transport permissions between various steelmaking and rolling workshops during a fire to prevent cascading accidents. In petrochemical industrial parks, this method can dynamically adjust the control permissions of access control, valves, and emergency broadcasts in affected areas based on real-time simulation results of toxic gas diffusion clouds, guiding personnel to evacuate safely. In intelligent warehousing and logistics centers, this method can dynamically authorize or restrict control of stacker cranes, conveyor belts, and fire shutters in specific aisles based on the fire spread situation, prioritizing the safety of high-value or hazardous materials warehouses.
[0119] The embodiments of this invention can be deployed as a standalone dynamic access control platform, integrating with existing enterprise fire alarm systems, industrial control systems (ICS), security monitoring systems (CCTV), and emergency command platforms via API interfaces; alternatively, they can be embedded as a core functional module into next-generation smart fire protection or industrial internet security platforms, providing them with dynamic access control capabilities based on the zero-trust principle. By liberating resource object permissions from pre-defined static attributes and dynamically binding them to real-time changing risk situations, this invention provides an effective technical path for solving the problem of safe and efficient management in the industrial field under extreme emergency conditions.
[0120] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.
Claims
1. A dynamic identity and permission adaptive control method under a zero-trust architecture, characterized in that, The method responds to an access request initiated by a user for a target resource object, and the method includes: The original physical address of the target resource object is converted into the target fire impact address, wherein the target fire impact address is determined based on the current fire propagation front and is used to characterize the ownership of the target resource object under the current fire propagation relationship; Based on the current identity verification result of the visitor, determine the visitor's personnel-side permissions for the target resource object; The device-side command permissions corresponding to the target resource object are determined based on the target fire impact address, and the personnel-side permissions are adjusted when both the personnel-side permissions and the device-side command permissions meet the control conditions corresponding to the access request.
2. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 1, characterized in that, Converting the original physical address of the target resource object to the target fire impact address includes: Based on the original physical address of the target resource object, determine the physical ownership unit of the target resource object in the industrial site; Based on the influence relationship between the current fire propagation front and the physical attribution unit, determine the fire influence unit currently corresponding to the target resource object; The target fire-affected address is obtained by re-addressing the original physical address of the target resource object according to the fire-affected unit.
3. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 2, characterized in that, Based on the influence relationship between the current fire propagation front and the physical attribution unit, the fire influence unit currently corresponding to the target resource object is determined, including: When the target resource object is located within the coverage area of the current fire propagation front, the directly affected unit corresponding to the current fire propagation front is determined as the fire impact unit; When the target resource object is located outside the coverage area of the current fire spread front, and the target resource object is on at least one of the paths of smoke diffusion path, heat radiation influence path, evacuation path or fire linkage path, the associated influence unit corresponding to the path is determined as the fire influence unit.
4. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 1, characterized in that, The target fire impact address includes at least one of the following: fire event identifier, fire impact unit identifier, object role identifier, address version information, and valid time information; The object role identifier is used to characterize the target resource object as belonging to one of the following categories under the current fire spread relationship: fire awareness object, evacuation impact object, smoke exhaust linkage object, hazard source associated object, fire water pressure guarantee object, or cross-regional control object.
5. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 1, characterized in that, Based on the current identity verification result of the visitor, determine the visitor's user-side permissions for the target resource object, including: The set of authorized fire scene addresses corresponding to the visitor is determined based on the current identity verification result. Determine whether the target fire-affected address falls within the set of authorized fire-affected addresses for personnel; If the target fire-affected address falls within the set of authorized fire-affected addresses, then the access personnel's personnel-side permissions to the target resource object are determined based on the request action corresponding to the access request.
6. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 5, characterized in that, The set of authorized fire scene addresses for personnel is determined based on the emergency task information, task area, task validity time, and identity verification status of the personnel visiting the site. The identity verification status is used at least to characterize whether the accessing person meets the continuous identity verification requirements under the zero-trust architecture when the access request occurs.
7. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 5, characterized in that, The method further includes: If the target fire impact address does not fall into the personnel fire authorized address set, then the fire impact unit corresponding to the target fire impact address is determined to be a fire impact unit corresponding to at least one authorized address in the personnel fire authorized address set; Based on the boundary association, the user-side permissions of the accessing personnel to the target resource object are adjusted, wherein the adjustment includes at least one of adjusting the control permission to the suggestion permission, adjusting the suggestion permission to the viewing permission, revoking the viewing permission, or rejecting the access request.
8. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 1, characterized in that, Determine the device-side command permissions corresponding to the target resource object based on the target fire-affected address, including: When the target resource object is a controllable device object, the control sources, control actions, and address versions that the controllable device object is allowed to accept, execute, and use under the current fire propagation relationship are determined based on the target fire impact address. The device-side command permissions are generated based on the control source, control action, and address version.
9. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 8, characterized in that, When both the personnel-side permissions and the device-side command permissions meet the control conditions corresponding to the access request, the personnel-side permissions are adjusted, including: Determine whether the user's terminal permissions allow the user to initiate the control action corresponding to the access request; Determine whether the device-side command permission allows the target resource object to accept the control source, control action, and address version corresponding to the access request; If all are allowed, the user's access permissions to the target resource object are adjusted to the executable permissions corresponding to the control action; If any of these permissions are not allowed, the user's access permissions to the target resource object will be adjusted to restricted permissions or denied permissions.
10. The dynamic identity and permission adaptive control method under a zero-trust architecture according to claim 1, characterized in that, The method further includes: When the current fire propagation front changes, the target fire impact address corresponding to the target resource object is re-determined; If the redefined target fire impact address is inconsistent with the original target fire impact address, the personnel terminal permissions and the device terminal command receiving permissions shall be updated based on the redefined target fire impact address. Specifically, when the target fire-affected address is within the address change stability window, the access personnel retain their viewing permissions for the target resource object, while restricting their access to the target resource pair.
