Power equipment potential intrusion target identification and active defense method and system

Abstract

The application relates to the technical field of target identification, and provides a power equipment potential invasion target identification and active defense method and system. In the method, the type of a potential invasion target is analyzed according to a visible light imaging diagram, an infrared light imaging diagram, a dielectric constant and an ice thickness; when the potential invasion target is multiple, the basic threat values of the potential invasion targets are calculated; the target threat values of the potential invasion targets are determined according to the basic threat values and the influence degrees among the potential invasion targets, the target threat values of the potential invasion targets are combined into a threat value sequence; a dynamically reconstructed defense chain is generated according to the threat value sequence and a preset defense means mapping rule, and corresponding active defense is carried out on different potential invasion targets according to the defense chain. The technical scheme provided by the application integrates multi-source data intelligent identification of power transmission line threats, realizes adaptive cooperative protection through a dynamic defense chain, and improves the reliability of active defense.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of target identification, and in particular to a power equipment potential threat target identification and active defense method and system. BACKGROUND

[0002] In the safe and stable operation of the power system, the identification and active defense of potential threat targets of power equipment play a crucial role. It can effectively reduce the failure rate, reduce the outage time, and ensure the continuity and reliability of power supply. The environment of the power transmission corridor is complex and changeable, and the main threats include suspended foreign objects, line mechanical overload caused by excessive ice, equipment external damage, and unauthorized personnel intrusion into critical areas. These threats are sudden, diverse, and have potential chain destruction risks. Therefore, there is an urgent need for a technical solution that can cooperatively perceive the multi-dimensional state information of the target area in a non-contact manner, intelligently analyze the specific threat type based on this, and actively take defensive measures according to the threat situation to maximize the safety of the power grid and the reliability of power supply.

[0003] The existing scheme is a visible light and infrared multi-sensor image fusion analysis system based on deep learning. The system is deployed on the power transmission line tower or inspection equipment, uses a high-definition visible light camera and an infrared thermal imager to collect image data of the target area; through a trained deep neural network model, the fused dual-spectrum image is automatically analyzed to identify abnormal targets and detect abnormal temperature rise points on the line based on infrared thermal images. Finally, the identification results are classified into preset threat categories.

[0004] However, the core defect of the existing scheme lies in the limitation of the perception dimension and the lack of threat quantification capability. First, visible light and infrared imaging mainly reflect the surface morphology and temperature distribution of the target, and it is difficult to directly and accurately obtain the physical properties and key quantitative indicators of the target, resulting in insufficient confidence in identifying certain threat types, being easily disturbed by environmental light, weather, target camouflage and other factors, and causing false positives or false negatives. Secondly, the existing scheme lacks a mechanism for threat degree quantification and evaluation of multiple threat targets identified; it can usually only output the threat type and location, and cannot calculate the "basic threat value" of each threat, nor can it consider the possible mutual influence relationship between multiple threat targets, so as to form a quantitative sequence representing the overall threat situation. This makes it difficult to dynamically optimize and accurately allocate resources according to the urgency and interrelation of threats in subsequent defense decision-making, and the defense response is often static or based on simple priority rules, making it difficult to achieve "dynamic reconstruction" of the cooperative defense chain. SUMMARY

[0005] The application provides a power equipment potential invasion target identification and active defense method and system to solve the low active defense reliability caused by the limited perception dimension and the lack of threat quantification capability in the prior art.

[0006] In a first aspect, the application provides a power equipment potential invasion target identification and active defense method, comprising:

[0007] Obtaining a visible light imaging image, an infrared light imaging image, a dielectric constant and an ice thickness of a target area, wherein the target area includes an aerial area where a power transmission line is located and a ground area below the power transmission line;

[0008] According to the visible light imaging image, the infrared light imaging image, the dielectric constant and the ice thickness, the type of the potential invasion target is analyzed, and the type is foreign matter on the power transmission line, excessive ice thickness, equipment external damage and personnel intrusion;

[0009] When the potential invasion target is multiple, the basic threat value of each potential invasion target is calculated;

[0010] According to the basic threat value, the influence degree between each potential invasion target is combined to determine the target threat value of each potential invasion target, and the target threat value of the potential invasion target is combined into a threat value sequence;

[0011] According to the threat value sequence, a dynamically reconstructed defense chain is generated according to a preset defense means mapping rule, and corresponding active defense is performed on different potential invasion targets according to the defense chain.

[0012] Optionally, the dynamically reconstructed defense chain is generated according to the threat value sequence and the preset defense means mapping rule, comprising:

[0013] According to the threat value sequence, the threat value interval to which the target threat value of each potential invasion target belongs is determined, and a defense unit set is generated according to the threat value interval to which the target threat value of each potential invasion target belongs and the corresponding defense resource type;

[0014] A topology constraint allocation operation is performed on the defense unit set to generate an initial defense chain;

[0015] A threat chain effect calculation and defense unit priority rearrangement operation between adjacent equipment is performed on the initial defense chain to generate a reconstructed defense chain.

[0016] Optionally, the threat chain effect calculation and defense unit priority rearrangement operation between adjacent equipment is performed on the initial defense chain to generate a reconstructed defense chain, comprising:

[0017] Based on the physical connection relationship of the defense unit associated equipment in the initial defense chain, threat propagation path calculation is performed to generate a threat propagation path;

[0018] The threat propagation path is subjected to threat intensity accumulation to generate a cumulative threat intensity;

[0019] Based on the cumulative threat intensity, the defense units in the initial defense chain are subjected to priority rearrangement to generate a reconstructed defense chain.

[0020] Optionally, the target threat value of each potential invasion target is determined according to the basic threat value and the influence degree between the potential invasion targets, the target threat values of the potential invasion targets are combined into a threat value sequence, which includes:

[0021] The influence degree between the potential invasion targets is subjected to conduction path analysis to generate a threat conduction path set;

[0022] The path weight of the threat conduction path set is calculated to generate a path weight set;

[0023] Based on the path weight set, the basic threat value of each potential invasion target is subjected to threat value correction to generate a target threat value;

[0024] The target threat values of all the potential invasion targets are subjected to device topology position sorting to generate a threat value sequence.

[0025] Optionally, the influence degree between the potential invasion targets is subjected to conduction path analysis to generate a threat conduction path set, which includes:

[0026] The connection strength of the connection topology between the potential invasion targets is quantified to form a connection strength set;

[0027] The connection strength set and the influence degree are associated by using a failure mode to generate a failure association set;

[0028] Based on the failure association set, a conduction path expansion operation is performed to generate an initial conduction path set;

[0029] The initial conduction path set is subjected to environmental interference correction to generate the threat conduction path set.

[0030] Optionally, the type of the potential invasion target is analyzed according to the visible light imaging image, the infrared light imaging image, the dielectric constant and the ice thickness, which includes:

[0031] The shape contour features in the visible light imaging image are extracted to obtain a contour feature set representing whether there is a foreign object on the power transmission line, whether the equipment is externally damaged and whether a person invades;

[0032] Perform a temperature anomaly region segmentation operation on the infrared light image to generate a set of hot spot features that characterize whether there are foreign objects or personnel intrusion on the transmission line;

[0033] Based on the dielectric constant, determine the set of node features that characterize the insulation degradation region where there is external equipment damage;

[0034] Perform ice damage risk classification on the ice thickness to generate a set of ice features that characterize whether there is an ice risk.

[0035] The contour feature set, the hot spot feature set, the node feature set, and the icing feature set are cross-fused to generate a multi-source feature spectrum;

[0036] Type matching is performed based on the multi-source feature spectrum to generate the type of potential attack targets.

[0037] Optionally, when there are multiple potential targets, calculating the basic threat value of each potential target includes:

[0038] Spatial threat quantification is performed on the spatial location parameters of each potential target to generate a corresponding spatial threat factor;

[0039] Extract the dynamic behavioral characteristics of each potential target and generate corresponding dynamic threat factors;

[0040] For potential targets such as foreign objects, excessive ice accumulation, and personnel intrusion on power transmission lines, the spatial threat factors and dynamic threat factors are fused to generate corresponding basic threat coefficients.

[0041] For potential targets of external damage to equipment, the material property parameters of the potential targets are used to calculate the attribute threat, generate attribute threat factors, and fuse the spatial threat factors, dynamic threat factors and attribute threat factors to generate the corresponding basic threat coefficient.

[0042] Secondly, this application provides a system for identifying and actively defending against potential threats to power equipment, comprising:

[0043] The acquisition module is used to acquire visible light imaging, infrared light imaging, dielectric constant and icing thickness of the target area, wherein the target area includes the air area where the transmission line is located and the ground area below the transmission line;

[0044] The analysis module is used to analyze the type of potential intrusion target based on the visible light imaging image, infrared light imaging image, dielectric constant and ice thickness, wherein the type is foreign object on the transmission line, excessive ice thickness, external equipment damage and personnel intrusion;

[0045] a calculating module, configured to calculate a basic threat value of each potential threat target when there are multiple potential threat targets;

[0046] a combining module, configured to determine a target threat value of each potential threat target according to the basic threat value and the influence degree between the potential threat targets, and combine the target threat values of the potential threat targets into a threat value sequence;

[0047] a generating module, configured to generate a dynamically reconstructed defense chain according to the threat value sequence and a preset defense means mapping rule, and perform corresponding active defense on different potential threat targets according to the defense chain.

[0048] In a third aspect, the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component, so as to realize the power equipment potential threat target identification and active defense method according to any one of the first aspect.

[0049] In a fourth aspect, the present application provides a computer storage medium, which stores a computer program; when the computer program is executed by a computer, the power equipment potential threat target identification and active defense method according to any one of the first aspect is realized.

[0050] In the present application, a power equipment potential threat target identification and active defense method is provided, which comprises the following steps: acquiring a visible light imaging image, an infrared light imaging image, a dielectric constant and an ice thickness of a target area, the target area including an air area where a power transmission line is located and a ground area below the power transmission line; analyzing a type of a potential threat target according to the visible light imaging image, the infrared light imaging image, the dielectric constant and the ice thickness, the type being foreign matter on the power transmission line, excessive ice, equipment external damage and personnel intrusion; calculating a basic threat value of each potential threat target when there are multiple potential threat targets; determining a target threat value of each potential threat target according to the basic threat value and the influence degree between the potential threat targets, and combining the target threat values of the potential threat targets into a threat value sequence; generating a dynamically reconstructed defense chain according to the threat value sequence and a preset defense means mapping rule, and performing corresponding active defense on different potential threat targets according to the defense chain.

[0051] The application synchronously acquires four-dimensional data of visible light imaging, infrared imaging, dielectric constant and ice thickness of the air and ground area of the power transmission line, realizes global multi-physical field collaborative perception, and improves the identification accuracy of four types of threats of foreign matter, excessive ice, equipment external damage and personnel intrusion; in the multiple threat coexistence scene, the threat level is dynamically corrected by calculating the basic threat value of each target and introducing the influence factor between targets, a quantitative associated threat value sequence is generated to accurately represent the composite threat situation; based on the sequence mapping the preset defense rule, the defense chain is dynamically reconstructed and differential active defense is implemented, multi-target collaborative disposal and adaptive optimization of defense resources are achieved, and the reliability of active defense for potential invasion targets is improved.

[0052] Further, the threat value interval to which each target threat value belongs is determined according to the threat value sequence, the defense resource type is mapped to generate a defense unit set, and the initial defense chain is generated by applying topological constraints to the set. The threat propagation path is calculated based on the physical connection relationship of the defense unit associated equipment, the cumulative threat intensity is obtained by accumulating the threat intensity, and the defense unit priority is rearranged to generate the reconstructed defense chain. The threat value interval matching realizes accurate adaptation of defense resources, and the topological constraint distribution guarantees the physical feasibility of the defense chain. By using the threat propagation path calculation and intensity accumulation mechanism, the key risk nodes are dynamically identified and the defense priority is rearranged, the multi-defense unit collaborative response and chain risk active inhibition are realized, and the defense efficiency in the composite threat scene is improved.

[0053] These aspects or other aspects of the application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0055] Figure 1 A flowchart of a power equipment potential invasion target identification and active defense method provided by an embodiment of the application;

[0056] Figure 2 A structural schematic diagram of a power equipment potential invasion target identification and active defense system provided by an embodiment of the application;

[0057] Figure 3 A structural schematic diagram of a computing device provided by an embodiment of the application. DETAILED DESCRIPTION

[0058] In order for those skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some of the processes described in the specification and claims of the present application and the accompanying drawings, multiple operations are described in a specific order. However, it should be clear that these operations can be performed in an order other than the order in which they are presented or in parallel, and the serial numbers of the operations, such as 11, 12, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this paper are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence. Also, "first" and "second" are different types.

[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0061] To solve the problem of low reliability of active defense caused by the limitation of perception dimension and the lack of threat quantification ability in the prior art, the present application provides a power equipment potential invasion target identification and active defense method. The method uses the following idea: to break through the single perception dimension and the discrete bottleneck of threat disposal of the traditional protection system, first, build an air-ground cooperative monitoring system, fuse four-dimensional heterogeneous data of visible light form, infrared temperature, dielectric constant material and ice thickness structure, and realize accurate identification of global threat with complementary enhancement of multiple physical fields; for multiple threat coexistence scenarios, innovatively design a threat quantification model, establish an objective evaluation benchmark through a basic threat value, introduce an influence factor between targets to dynamically correct the threat level, and generate a sequenced threat list to represent the complex risk situation; based on the sequence mapping preset rule, dynamically arrange the defense unit, realize multi-target adaptive cooperative protection through real-time reconstruction of the defense chain, and form a closed-loop intelligent response mechanism from perception, evaluation to defense.

[0062] Figure 1 A flowchart of a power equipment potential invasion target identification and active defense method provided by the embodiments of the present application is shown in Figure 1 The method comprises the following steps.

[0063] S11. Obtain visible light imaging, infrared imaging, dielectric constant and icing thickness of the target area. The target area includes the air area where the transmission line is located and the ground area below the transmission line.

[0064] The target area can refer to the three-dimensional monitoring range including the space where the transmission line is suspended and the passageway beneath the towers, where multi-source sensing devices can be deployed. Visible light imaging refers to electromagnetic wave images captured by cameras, used to reflect the target's shape and surface texture. Infrared imaging can be infrared radiation images acquired by thermal imagers, used to characterize the target's temperature distribution. Dielectric constant refers to the dielectric properties of the line insulation material in an alternating electric field, obtained through measurement by a capacitance sensor. Icing thickness refers to the maximum vertical dimension of the ice layer on the line surface, calculated using laser scanning point cloud data fitting. A transmission line can include an overhead conductor, insulator strings, towers, and fittings forming the power transmission channel. The aerial area refers to the space where the conductor is suspended at a specified distance from the ground, primarily monitoring foreign object attachment and icing conditions. The ground area can refer to a designated area around the tower foundation, focusing on detecting personnel intrusion and equipment damage.

[0065] In this embodiment, firstly, visible light and infrared light images of the air area where the transmission line is located are collected by a UAV-borne multispectral device, and at the same time, the ground monitoring station is used to obtain the same source data of the ground area below the transmission line; secondly, the dielectric constant of the insulation is measured in real time by a dielectric sensor embedded in the conductor, and the ice thickness of the conductor is scanned by a laser ranging device, and finally a four-dimensional monitoring dataset covering the entire air and ground area is formed.

[0066] S12. Based on the visible light imaging image, infrared light imaging image, dielectric constant, and ice thickness, the type of potential intrusion target is analyzed. The types are foreign objects on the power transmission line, excessive ice thickness, external equipment damage, and personnel intrusion.

[0067] Among them, potential targets of harm refer to four types of risk entities identified through multi-source data analysis. Foreign objects can refer to plastic bags, flocks of birds, etc. Excessive ice thickness indicates that the thickness exceeds a preset thickness threshold, and external equipment damage indicates that the insulation on the equipment surface has cracks or damage. Personnel intrusion can be understood as unauthorized personnel appearing in a prohibited area and performing some action within the prohibited area, such as digging or climbing poles.

[0068] In this embodiment, the outline of foreign objects and the characteristics of equipment damage are first extracted from the visible light imaging image; then, the infrared light imaging image is processed using thermal image segmentation technology to identify hot spots of personnel intrusion; then, the insulation degradation area is judged based on the dielectric constant threshold, and the ice damage risk points are marked by combining the ice thickness grading model; finally, the outline features, hot spot features, dielectric features and ice features are integrated through a multimodal feature fusion network to output type labels for four types of potential targets of harm on the transmission line: foreign objects, excessive ice thickness, external equipment damage and personnel intrusion.

[0069] S13. When there are multiple potential targets, calculate the base threat value of each potential target.

[0070] Among them, the basic threat value refers to the initial risk quantification value calculated based on the target's spatial location and dynamic behavior.

[0071] In this embodiment, when multiple potential threats are identified, a spatial threat quantification algorithm is first applied to the spatial location parameters of each target to generate a spatial threat factor; then, dynamic behavioral characteristics of the target are extracted to generate a dynamic threat factor; subsequently, for targets such as foreign objects, excessively thick ice, and personnel intrusion, the spatial threat factor and the dynamic threat factor are fused to generate a basic threat coefficient; for targets involving external damage to equipment, attribute threat calculation is performed on their material property parameters to generate an attribute threat factor, and the three types of factors are fused to generate a basic threat coefficient, finally outputting a set of basic threat values ​​for all targets.

[0072] S14. Based on the basic threat value and the degree of influence between each potential target, determine the target threat value of each potential target and combine the target threat values ​​of the potential targets into a threat value sequence.

[0073] The degree of impact refers to the risk transmission intensity coefficient between targets through physical connections of equipment. The target threat value is the comprehensive risk quantification value after correction for the degree of impact of associated targets. The threat value sequence is the ordered set of target threat values ​​sorted according to the topological location of the equipment.

[0074] In this embodiment, firstly, the connection topology strength between potential targets is quantified to form a connection strength set; secondly, the connection strength and the degree of impact are correlated with the failure mode library to generate a failure correlation set; then, an initial transmission path set is generated through a transmission path expansion algorithm, and an environmental interference correction factor is introduced to optimize path reliability; finally, the path weights of the threat transmission path set are calculated, and the basic threat values ​​are linearly weighted and corrected based on the weights to generate target threat values, which are then sorted by device topology location to form a threat value sequence.

[0075] S15. Based on the threat value sequence and the preset defense method mapping rules, generate a dynamically reconstructed defense chain, and carry out corresponding active defense against different potential targets based on the defense chain.

[0076] Among them, the preset defense method mapping rules refer to the database that defines the matching relationship between threat value ranges and defense resources such as laser de-icing or sonic repellency. The defense chain refers to the deployment and activation sequence of defense units according to execution priority. Active defense can refer to physical intervention measures automatically triggered by the system, including actions such as de-icing, repellency, and maintenance.

[0077] In this embodiment, the threat value sequence is first mapped to a preset defense rule base, and defense resource types such as laser de-icing devices and acoustic wave de-icing devices are matched to generate a set of defense units. Secondly, the location allocation of defense units is performed based on the transmission line spatial topology model to construct an initial defense chain. Then, the intensity of the chain effect between adjacent devices is calculated through a fault propagation model, and the defense priority is rearranged according to the cumulative threat intensity. Finally, the reconstructed defense chain is generated and active defense measures such as drone cruise de-icing and directional acoustic wave warning are activated.

[0078] Here is a specific example: First, visible light and infrared images of spans 3-5 of a 220kV line are acquired using a drone, while insulator dielectric constant and conductor icing data are simultaneously transmitted back by tower sensors. Next, a convolutional neural network is used to identify the outline of a kite-like foreign object and the hot spots of construction workers below in the images. Combined with abnormal dielectric constants, insulator damage is determined, and icing damage points are marked based on icing thickness exceeding a threshold. Then, the spatial threat factor of the kite's distance from the conductor and the dynamic threat factor of its swing amplitude are calculated, and a weighted basic threat value is generated. The impact of the construction crane on adjacent insulators is quantified, and the target threat value is corrected and a sequence is generated. Next, laser de-icing devices and acoustic wave de-icing devices are matched according to defense rules and allocated to tower 4 and the ground area according to the line topology to form an initial defense chain. Finally, it is analyzed that icing on tower 4 may lead to cascading failures on adjacent towers. Based on the cumulative threat intensity, de-icing operations are initiated first, and acoustic wave de-icing warnings are triggered simultaneously, completing the dynamic defense response.

[0079] By executing S11~S15, this embodiment of the application achieves accurate identification of complex threats by integrating four-dimensional perception data of air and ground areas, establishes a multi-target associated threat assessment model and a dynamic defense chain generation mechanism, and effectively improves the power transmission system's ability to coordinate the handling of complex risks and the efficiency of defense resource utilization.

[0080] In one possible embodiment, S15, based on the threat value sequence and combined with preset defense measure mapping rules, a dynamically reconstructed defense chain is generated, including:

[0081] Step 151: Based on the threat value sequence, determine the threat value range to which the target threat value of each potential intrusion target belongs, and generate a set of defense units based on the threat value range to which the target threat value of each potential intrusion target belongs and the corresponding defense resource type.

[0082] The threat value range refers to a continuous range of target threat values ​​defined based on historical fault data, used to distinguish threat levels, including three decision thresholds: low threat, medium threat, and high threat. The defense resource type refers to a classification of deployable physical protective equipment, including four types of physical resources: laser de-icing devices, acoustic deterrents, drone inspection units, and mechanical maintenance robots. The defense unit set refers to an instantiated combination of defense resources scheduled to address multiple threat targets, including resource type identifiers, deployment quantities, and expected location parameters.

[0083] In this embodiment, the target threat value of each potential intrusion target in the threat value sequence is first compared with multiple preset continuous numerical ranges to determine its specific threat value interval; then, based on the pre-established table of the correspondence between threat value intervals and defense resource types, the corresponding defense resource type is matched for each target; finally, the defense resource types and their instantiation quantities of all targets are aggregated to generate a set of defense units.

[0084] Step 152: Perform a topology constraint assignment operation on the set of defense units to generate an initial defense chain.

[0085] Among them, the topology constraint allocation operation can refer to a resource optimization algorithm that, based on the spatial topology map of transmission lines, restricts the deployment locations of defense units to physically accessible areas of equipment while satisfying minimum electrical safety distances. The initial defense chain can refer to a linear execution sequence of defense units generated only considering spatial feasibility, and its order is based on the high and low levels of the threat value intervals corresponding to the defense units.

[0086] In this embodiment, the spatial topology model of the transmission line is first loaded; then, based on the functional characteristics of each unit in the defense unit set, the Dijkstra algorithm is used to calculate its optimal deployment path on the transmission equipment; subsequently, the spatial location allocation of the defense units is performed according to the electrical safety distance constraints and mechanical connection relationships between the equipment; finally, an initial defense chain that satisfies the physical topology feasibility is generated.

[0087] Step 153: Perform threat chain effect calculation between adjacent devices and priority reordering of defense units on the initial defense chain to generate a reconstructed defense chain.

[0088] Adjacent equipment refers to pairs of equipment nodes in a power transmission system that have direct electrical connections or mechanical linkages, such as the connection point between an insulator string and a conductor, or adjacent towers within the same tension section. Threat cascading effect calculation refers to quantifying the additional risk intensity caused to physically related equipment when a device fails, using a fault propagation model. The calculation method is to multiply the base threat intensity by the coupling coefficient of the connection topology, and then divide by the equipment aging coefficient. Defense unit priority reordering refers to the optimization process of readjusting the execution order of the defense chain based on the cumulative threat intensity matrix, ensuring that high-cascading-risk equipment receives priority in handling resources. Reconstructing the defense chain refers to the final defense sequence after integrating spatial topology constraints and cascading effect optimization; its elements are a set of defense unit deployment instructions ordered according to the new priority.

[0089] In this embodiment of the application, in the detailed process of the threat chain effect in step 153, the chain risk value between adjacent defense units in the initial defense chain is first calculated. The specific formula is chain risk value = upstream device threat value / downstream device threat value × device connection strength coefficient. When the chain risk value is greater than a preset threshold, it is determined that there is a chain effect. Secondly, the priority of defense units is rearranged according to the chain risk value. For the device group with chain effect, the execution order of the defense unit is promoted to before the device without chain effect. Finally, in the generated reconstructed defense chain, the defense operation of the high chain risk device group will be executed first, thereby blocking the threat transmission path.

[0090] In this embodiment, firstly, the pairs of adjacent device nodes with direct electrical connections in the initial defense chain are analyzed; secondly, the threat transmission intensity to associated devices when a single device fails is simulated using a fault tree model, where the intensity value is the base threat intensity multiplied by the coupling coefficient of the connection topology; then, the cascading effect intensity on the transmission path is weighted and accumulated by time decay to generate a cumulative threat intensity matrix; finally, the execution order of the defense units is rearranged in descending order according to the matrix values, and a dynamically optimized reconstructed defense chain is output.

[0091] Here is a specific example: First, a drone was used to capture visible light and infrared images of towers 3 to 5 of a 220kV transmission line, simultaneously receiving insulator performance data and conductor icing thickness information transmitted from tower sensors. Second, intelligent recognition technology detected the abnormal shape of a kite hanging from tower 3 and detected a human heat source signal in the construction area below tower 4. Based on the abnormal insulation data, damage to the insulators was determined. Simultaneously, the measured icing thickness exceeding safety standards indicated a potential ice hazard at tower 4. Subsequently, threat quantification calculations were performed: measuring the distance between the kite and the conductor at 0.5 meters yielded a positional threat factor of 20, calculated by multiplying the reciprocal of the distance by 100; combining this with the kite's 0.2-meter swing amplitude, a dynamic threat factor of 10 was obtained, calculated by multiplying the swing amplitude by 50; averaging the two factors yielded a base threat value of 15. Simultaneously assess the impact of the construction crane on surrounding equipment, multiply the base value of 15 by a location impact coefficient of 1.2, and finally generate a kite target threat value of 18, forming a descending threat sequence that includes an ice disaster point threat value of 12 and a construction worker threat value of 9.

[0092] Next, defense measures were matched according to the threat level: laser elimination equipment was configured for high-threat targets, acoustic warning devices for medium-threat targets, and monitoring equipment for low-threat targets. Based on the actual connection method of the transmission line, laser equipment was deployed on tower No. 3 to deal with kite threats, acoustic devices were installed on the ground area of ​​tower No. 4 to drive away construction workers, and monitoring equipment was set up on tower No. 5, thus constructing the initial defense system.

[0093] Finally, the defensive linkage effect was analyzed, and the impact of icing on tower No. 4 on adjacent towers was calculated. The linkage risk value was calculated by dividing the current threat value (12) by the threat value of adjacent equipment (9), and then multiplying by the equipment connection strength coefficient (0.9). Based on this value, the defense priority was readjusted: de-icing operations on tower No. 4 were initiated first, simultaneously triggering the laser de-icing device and acoustic warning system on tower No. 3, achieving coordinated protection across the entire line. Throughout the process, threat value calculations used physical measurement data of distance and sway amplitude, defense deployment strictly followed the physical location relationships of the equipment, and risk linkage assessment was based on equipment connection characteristics, ensuring the scientific validity and effectiveness of the defense measures.

[0094] By executing steps 151 to 153, this embodiment of the application accurately matches the type of defense resources with the threat level, generates an executable deployment scheme based on spatial topology constraints, dynamically optimizes the defense priority by combining the risks of device cascading, effectively blocks the fault propagation path of multiple devices, and improves the collaborative response efficiency of the defense system.

[0095] In one possible embodiment, step 153, performing threat cascading effect calculations between adjacent devices and prioritizing defense units on the initial defense chain to generate a reconstructed defense chain, includes:

[0096] Step a1: Based on the physical connection relationship of the defense unit associated devices in the initial defense chain, calculate the threat propagation path and generate the threat propagation path.

[0097] Among them, defense unit associated devices refer to hardware or software components in the network or system that are directly or indirectly connected to the security defense unit, including routers, firewalls, and intrusion detection systems, used to implement security functions and obtained based on the initial defense chain configuration. Physical connection relationships refer to the linking methods between devices based on actual hardware or network topology, including cable connections, wireless channels, or interface bonding, used to reflect communication dependencies between devices and obtained based on network architecture analysis. Threat propagation paths refer to the specific propagation sequence of a threat from its source to the target device, including device nodes and connection edges, used to represent the direction of attack spread and calculated based on the threat propagation path.

[0098] In this embodiment, firstly, associated devices of the defense units are extracted from the initial defense chain. These devices include security components such as firewalls and routers. Secondly, the physical connections between these associated devices, such as network topology links, are analyzed. Subsequently, based on the physical connections, graph theory algorithms, such as depth-first search, are used to calculate threat propagation paths and identify all possible threat paths. Finally, a set of threat propagation paths is generated as the output.

[0099] Step a2: Accumulate the threat intensity along the threat propagation path to generate the cumulative threat intensity.

[0100] Threat intensity aggregation refers to the aggregation operation of threat values ​​at various points along the path, including addition or weighted summation, used to quantify the overall threat level, and is obtained based on the intensity values ​​along the threat propagation path. Cumulative threat intensity is the comprehensive value after aggregating the threat intensities of all threats along the path, including numerical scores or risk indicators, used to assess the severity of the path, and is generated based on the threat intensity aggregation operation.

[0101] In this embodiment, firstly, the threat propagation paths generated in step a1 are obtained. Secondly, for each threat propagation path, threat strength values, such as attack probability or vulnerability scores, are extracted. Subsequently, an addition algorithm is used to accumulate the threat strength values ​​along the path. Finally, cumulative threat strength data for each path is generated as input for the next step.

[0102] Step a3: Based on the cumulative threat intensity, prioritize the defense units in the initial defense chain to generate a reconstructed defense chain.

[0103] Priority reordering refers to adjusting the processing order of defense units, including using sorting algorithms to rearrange them according to key indicators to optimize resource allocation, based on cumulative threat intensity analysis.

[0104] In this embodiment, firstly, the impact of cumulative threat strength on the defense units in the initial defense chain is analyzed. Secondly, a sorting algorithm, such as quicksort, is used to prioritize the defense units according to their cumulative threat strength values, adjusting their security response order. Subsequently, the reordering results are integrated. Finally, a reconstructed defense chain is generated, in which the defense units are arranged according to the new priority.

[0105] Here is a specific example: First, threat propagation analysis is conducted based on the initial defense system already established for the 220kV line from towers 3 to 5. This system includes laser removal equipment deployed on tower 3 to eliminate kite threats, an acoustic repellent device configured on tower 4 to handle icing hazards from construction workers and de-icing equipment, and monitoring equipment installed on tower 5. These devices are physically connected via the power line to form a chain structure. The direction of threat propagation is identified using a path analysis algorithm: when a kite with a target threat value of 18 appears on tower 3, the threat will propagate along the conductor to tower 4; simultaneously, the icing threat value of 12 on tower 4 will spread to adjacent equipment, thus generating two critical propagation paths: the first path propagates from tower 3 through the line to tower 4 and then to tower 5; the second path propagates directly from tower 4 to tower 5. Next, the threat intensity of each path is calculated cumulatively: the cumulative intensity of the first path is 35, calculated by adding threat value 18 (threat value 3), threat value 12 (threat value 4), and base threat value 5 (threat value 5); the cumulative intensity of the second path is 17, calculated by adding threat value 12 (threat value 4) and threat value 5 (threat value 5). Finally, the defense priority is adjusted based on the cumulative intensity values: the original execution order was laser clearing (threat value 3) first, followed by de-icing (threat value 4), and then monitoring (threat value 5). Since the first path has the highest cumulative value of 35, laser device 3 at the path's starting point is upgraded to the highest response level; the second path has the next highest cumulative value of 17, and de-icing device 4 remains a secondary priority. The reconstructed defense chain execution order is: immediately activate laser device 3 to clear kite threats, simultaneously activate de-icing device 4 to eliminate icing hazards, maintain monitoring device 5 operation, and continuously run sonic devices in the area of ​​tower 4 to drive away construction personnel. The entire process dynamically optimizes the allocation of defense resources by quantifying the risks of transmission paths, ensuring that high-threat transmission paths are given priority in handling. Threat value calculation strictly uses physical measurement data, path analysis is based on the actual connection relationship of equipment, and priority adjustment is based on the cumulative intensity formula to achieve scientific decision-making.

[0106] By executing steps a1 to a2, the adaptability and response efficiency of the embodiments of this application are improved, the risk of potential threat spread is effectively reduced, and the overall resilience of the system is enhanced.

[0107] In one possible embodiment, S14, based on the base threat value and considering the degree of influence between each potential target, determine the target threat value of each potential target, and combine the target threat values ​​of the potential targets into a threat value sequence, including:

[0108] Step 141: Analyze the transmission path of the impact between potential targets to generate a set of threat transmission paths.

[0109] Transmission path analysis refers to identifying the links in the threat propagation process between targets. This includes using graph theory algorithms to analyze device dependencies, simulating the direction of threat propagation, and obtaining the result based on the target association topology. The threat propagation path set is the collection of all threat propagation paths derived from the analysis, including node connection sequences and directional attributes. It describes the complete propagation network and is generated based on the transmission path analysis.

[0110] In this embodiment, the threat propagation path in step a1 and the threat transmission path in step 141 have an inheritance and extension relationship. The threat propagation path is based on the physical connection relationship between the defense unit and the associated devices. A graph traversal algorithm is used to generate a basic path describing the physical spread of threats between devices. This path is used as the underlying data input to step 141. Step 141 then analyzes the functional impact between potential targets based on this propagation path. By superimposing impact weights, the original propagation path is transformed into an enhanced transmission path. Other newly added impact links are integrated to form the threat transmission path set in step 141, realizing the upgrade from physical diffusion routes to a multi-dimensional impact network.

[0111] In this embodiment, firstly, attribute data and interrelationships of each potential target are acquired. Secondly, a graph traversal algorithm is used to analyze the impact transmission patterns between targets and identify the links in the threat spread. Subsequently, path modeling technology is used to perform topological connection analysis on the transmission relationships to clarify the direction and level of threat transmission. Finally, all transmission links are integrated to generate a set of threat transmission paths containing multiple sets of node sequences.

[0112] Step 142: Calculate the path weights of the threat propagation path set and generate a path weight set.

[0113] Path weight refers to the quantified threat impact of a single transmission path, including a combined calculated value of length factor and equipment risk coefficient, used to assess path criticality, and derived based on path attributes and a weight model. The path weight set is a dataset of weights for all transmission paths, including a unique identifier and weight value for each path, used for global threat correction, and generated based on path weight calculations.

[0114] In this embodiment, firstly, each path and its associated device attributes are extracted from the threat propagation path set. Secondly, a weight calculation model is designed based on path length, device vulnerability level, and connection criticality. Subsequently, a weighted algorithm is used to multiply the path length by the device risk factor and then divide by the path stability coefficient to calculate the quantified value for each path. Finally, the weight value of each path is output to form a path weight set.

[0115] Step 143: Based on the path weight set, adjust the basic threat value of each potential target to generate the target threat value.

[0116] Threat value correction refers to the process of adjusting the target's base threat value based on path weights, including multiplicative correction and normalization, to reflect the transmission and superposition effect, and is obtained based on the path weight set and the target's base value.

[0117] In this embodiment, firstly, the basic threat value of each potential target and the path weight set generated in step 142 are read. Secondly, a correction coefficient is designed based on the weight value of the path where the target is located, and the basic threat value is multiplied by the weight coefficient of the associated path. Subsequently, for targets with multiple associated paths, normalization correction is performed using the maximum weight value or the average weight value. Finally, a target threat value dataset reflecting the impact of transmitted threats is generated.

[0118] Step 144: Sort the target threat values ​​of all potential targets by device topology location to generate a threat value sequence.

[0119] Among them, device topology location sorting refers to sorting threat values ​​according to the spatial hierarchy of devices in the network, including core priority or proximity priority rules, to optimize the threat response order, based on target threat values ​​and topology coordinates.

[0120] In this embodiment, the difference between device topology location sorting and priority reordering is as follows: device topology location sorting is a static sorting of target threat values ​​based on the physical layer of the device in the network, considering only spatial location attributes, and generating a fixed sequence in descending order of coreness; priority reordering is based on the dynamic chain effect of threat propagation, and adjusts the execution order of defense units in real time through the formula chain risk value = upstream threat value / downstream threat value × connection strength coefficient, forming a reconstructed defense chain to respond to the risk of transmission.

[0121] In this embodiment, firstly, the target threat values ​​of all potential targets and their location coordinates in the network topology are obtained. Secondly, location priority rules are constructed based on the device's hierarchical depth and proximity in the topology. Subsequently, a spatial sorting algorithm is used to arrange the target threat values ​​in descending order from the core layer to the edge layer. Finally, a threat value sequence that integrates the topology location relationships is generated.

[0122] Here is a specific example: First, a threat propagation path analysis was performed on the section of the 220kV line from towers 3 to 5. Three critical paths were identified through the equipment connection topology: the first path propagates from the kite threat point on tower 3 to tower 4 via the conductor; the second path spreads from the icing point on tower 4 to the adjacent tower 5; and the third path propagates from the construction area on tower 4 to the control center. Next, the path weights were calculated: based on the number of insulator damage points along the path multiplied by the line distance coefficient. For example, the path from tower 3 to tower 4 has 2 damage points and a distance of 50 meters, so the weight = 2 × (1 / 50) × 100 = 4; the path from tower 4 to tower 5 has 1 damage point and a distance of 30 meters, so the weight = 1 × (1 / 30) × 100 ≈ 3.33; the path in the construction area has no damage but a distance of 80 meters, so the weight = 0.8, forming a weight set [4, 3.33, 0.8]. Next, the threat values ​​were adjusted: the threat value of the icing foundation of tower 4 (12) was multiplied by the associated maximum path weight of 4, generating a target threat value of 48; the threat value of the kite on tower 3 (18) was multiplied by the weight of 4, resulting in 72; the threat value of the construction area (9) was multiplied by the weight of 0.8, resulting in 7.2. Then, the threats were sorted by equipment level: tower 4 (48), located in the area control center, was ranked first, followed by tower 3 (72), and tower 5 (7.2). The final threat sequence [4:48, 3:72, 5:7.2] was generated, guiding the priority allocation of a laser de-icing device to tower 4, the configuration of an auxiliary de-icing device to tower 3, and the deployment of monitoring equipment to tower 5. The entire path analysis was based on the physical connection of the lines, the weight calculation used the formula of dividing the number of damaged points by the distance, the threat adjustment reflected the transmission and superposition effect, and the sorting followed the principle of prioritizing the control core.

[0123] By executing steps 141 to 144, this application embodiment quantifies the impact weight of threat transmission paths, dynamically corrects target threat values, and integrates device topology location relationships to achieve more accurate threat assessment and response priority allocation, effectively improving the protection targeting of key targets and reducing systemic security risks.

[0124] In one possible embodiment, step 141 involves performing a transmission path analysis on the degree of influence between potential targets to generate a set of threat transmission paths, including:

[0125] Step c1: Quantify the connection strength of the connection topology between each potential intrusion target to form a connection strength set.

[0126] In this context, connection topology refers to the layout structure of physical or logical connections between devices, including bus, star, or mesh architectures, used to reflect the communication relationships between devices, and obtained based on network deployment and configuration. Connection strength is a quantitative indicator of the stability of a single topology link, including a combined value of data transmission success rate and fault tolerance, used to evaluate connection reliability, and obtained based on link performance testing. The connection strength set is a dataset of connection strengths for all devices, including source and destination targets and strength values, used for global failure analysis, and generated based on connection topology quantification.

[0127] In this embodiment, firstly, the device attributes and topological connections of each potential target are obtained. Secondly, the communication frequency and data transmission reliability between devices are quantified through communication traffic analysis or signal transmission stability testing. Subsequently, the comprehensive connection strength value of each topological link is calculated by combining the connection medium type and link redundancy mechanism. Finally, the connection strength data of all device pairs are integrated to form a connection strength set.

[0128] Step c2: Using failure modes, correlate the set of connection strengths and the degree of impact to generate a failure correlation set.

[0129] Among them, failure modes refer to typical scenarios in which equipment failures cause conduction interruptions, including hardware damage, signal interference, or protocol errors. These are used to predict the point of interruption in threat propagation and are extracted from a historical failure database. The failure association set is a table showing the correspondence between connection strength and impact, including strength thresholds, failure probabilities, and impact coefficients. It is used for path expansion analysis and is obtained based on failure mode mapping.

[0130] In this embodiment, the process of generating the degree of impact can be described as follows: First, key parameters of equipment function are extracted based on the historical fault database. The degree of impact is calculated using the weighted formula: Degree of impact = Equipment function weight × Influence range coefficient. The equipment function weight is based on its allocation at the control level in the power system, and the influence range coefficient is calculated by dividing the number of households affected by power outages or the number of associated equipment by a baseline value of 1000. Second, dynamic correction is made in conjunction with real-time operating status. For example, when the equipment load rate exceeds 85%, it is multiplied by an overload coefficient of 1.2. Finally, a quantified dataset of the degree of impact is generated as the basis for the correlation between failure modes and connection strength.

[0131] In this embodiment, firstly, a predefined failure mode library is loaded to extract equipment failure types and propagation characteristics. Secondly, the strength values ​​in the connection strength set are mapped to the degree of influence between targets to establish a strength-failure probability correspondence. Subsequently, an association rule algorithm is used to analyze the amplification effect of connection strength attenuation on the degree of influence. Finally, a failure association set containing a ternary relationship of strength-failure-influence is generated.

[0132] Step c3: Perform a propagation path expansion operation based on the failure association set to generate an initial propagation path set.

[0133] The initial transmission path set refers to the expanded set of basic transmission paths, including direct paths and newly added indirect paths, which are used for environmental correction and are generated based on the transmission path expansion operation.

[0134] In this embodiment, firstly, direct propagation paths are identified based on the failure association set. Secondly, based on the amplification effect in the failure associations, indirect connection paths are added to expand the original propagation network. Subsequently, a path discovery algorithm is used to traverse all possible propagation directions and merge the newly added paths. Finally, an initial propagation path set containing both direct and indirect propagation paths is generated.

[0135] Step c4: Correct the initial transmission path set for environmental interference to generate a threat transmission path set.

[0136] Among them, environmental disturbance correction refers to the process of adjusting path stability based on environmental factors, including interference factor weighting and regional attenuation superposition, which is used to improve path authenticity and is based on environmental monitoring data and path set.

[0137] In this embodiment, firstly, environmental interference factor data, including electromagnetic interference intensity or temperature and humidity fluctuation values, are collected. Secondly, an interference factor model is established, multiplying the stability of each path in the initial transmission path set by an interference attenuation coefficient. Subsequently, the interference weights of cross-regional paths are adjusted a second time. Finally, a set of threat transmission paths optimized for environmental adaptability is output.

[0138] Here is a specific example: First, a physical connection analysis was performed on the 220kV line from towers 3 to 5. By measuring conductor resistance and signal transmission success rate, the connection strength between towers was quantified: the connection strength value between towers 3 and 4 was 0.92, and the connection strength value between towers 4 and 5 was 0.88, forming a set of connection strengths. Next, based on the icing failure mode, the connection strength was correlated with the impact range of the line break: the connection strength of 0.92 between towers 3 and 4 corresponded to an impact coefficient of 1.5, and the strength of 0.88 between towers 4 and 5 corresponded to a coefficient of 1.2, generating a set of failure correlations. Then, based on the failure correlations, the transmission path was expanded: the original direct path "from tower 3 to tower 4 icing point" was supplemented with an indirect path via tower 2, "from tower 3 to tower 2 to tower 4," forming an initial set of transmission paths. Subsequently, meteorological environmental data was collected, and a humidity attenuation coefficient of 0.85 was applied to the path crossing mountainous areas to correct transmission stability. Finally, a set of threat transmission paths adapted to terrain and climate was output, guiding the priority deployment of dual de-icing devices on the path between towers 3 and 4, and the installation of monitoring equipment on tower 2. The connection strength throughout the process is calculated based on measured resistance and communication data. Failure correlation adopts the ice thickness threshold and the influence range mapping formula. Path expansion is based on the terrain meandering characteristics. Environmental correction incorporates real-time meteorological parameters.

[0139] By executing steps c1 to c4, this embodiment of the application quantifies the dynamic correlation between connection strength and failure mode, and performs dual correction of the transmission path by combining environmental interference factors, which significantly improves the accuracy and scenario adaptability of threat transmission analysis and effectively enhances the pertinence of critical equipment protection strategies.

[0140] In one possible embodiment, S12, based on the visible light image, infrared image, dielectric constant, and icing thickness, the type of potential target is determined, including:

[0141] Step 121: Extract the shape and contour features from the visible light image to obtain a set of contour features that characterize whether there are foreign objects, damaged equipment, or unauthorized personnel on the power transmission line.

[0142] Among them, shape contour features refer to the quantitative description of the geometric properties of an object's boundary, including edge curvature, number of corners, and closure, used to identify physical anomalies, and are obtained based on visible light image analysis. The contour feature set refers to the integrated dataset of all contour features, including position coordinates and feature vectors, used for multi-source fusion, and is generated based on shape contour feature extraction.

[0143] In this embodiment, the visible light image is first processed using an image edge detection algorithm to extract the geometric boundary information of the power transmission line and surrounding objects. Secondly, shape contour features, including line segment curvature and the integrity of closed regions, are analyzed to identify abnormal protrusions or fracture patterns. Subsequently, features characterizing foreign object attachment, equipment casing damage, and human body contours are integrated into a contour feature set.

[0144] Step 122: Perform temperature anomaly region segmentation on the infrared light image to generate a set of hot spot features that characterize whether there are foreign objects or personnel intrusion on the transmission line.

[0145] The temperature anomaly region segmentation operation refers to the process of separating overheated regions in infrared images, including threshold segmentation and region merging, used to locate thermal anomalies, and is obtained based on infrared image processing. The hotspot feature set refers to the attribute dataset of the temperature anomaly region, including hotspot area, peak temperature, and gradient distribution, used to distinguish heat source types, and is generated based on the temperature anomaly region segmentation.

[0146] In this embodiment, a region growing algorithm is first used to segment the temperature field in the infrared image to identify abnormal temperature zones above the environmental threshold. Secondly, hotspot features, including temperature gradient distribution and area proportion, are extracted to distinguish between equipment overheating and human heat sources. Finally, a set of hotspot features characterizing foreign object heating or intruder thermal signals is generated.

[0147] Step 123: Based on the dielectric constant, determine the set of node features that characterize the insulation degradation area where there is external equipment damage.

[0148] Among them, the insulation degradation area refers to a localized part of the equipment where the dielectric properties deteriorate, including cracked areas or damp areas, used for early warning of external equipment failure, and is obtained based on dielectric constant detection. The node feature set refers to a quantitative dataset of the insulation degradation area, including location identifiers and degradation indices, used for risk assessment, and is generated based on dielectric characteristic analysis.

[0149] In this embodiment, the dielectric loss value of the power transmission and transformation equipment is first obtained using a dielectric sensor. Secondly, the insulation degradation area is located based on the change in dielectric constant, and the degradation degree index is calculated. Finally, a set of node features reflecting the risk of external damage to the equipment is generated.

[0150] Step 124: Perform ice damage risk classification on the ice thickness to generate a set of ice accretion features that characterize whether there is an ice accretion risk.

[0151] The ice damage risk classification operation refers to the process of dividing risk levels according to ice thickness, including setting interval thresholds and level labeling, used for ice disaster early warning, and based on ice thickness measurements. The ice feature set refers to the dataset of ice conditions, including ice thickness values, risk levels, and coverage areas, used for fusion analysis, and generated based on the ice damage risk classification.

[0152] In this embodiment, the thickness of the ice accretion on the conductor is first calculated using laser ranging data. Then, the ice damage is classified into light, moderate, and severe levels based on the thickness range. Finally, a set of ice accretion features containing the location and risk level of the ice accretion is generated.

[0153] Step 125: Perform feature cross-fusion on the contour feature set, hot spot feature set, node feature set and icing feature set to generate a multi-source feature spectrum.

[0154] Feature cross-fusion refers to the process of associating and integrating multimodal features, including feature concatenation and weight allocation, used to enhance discriminative information, and is obtained based on multi-source feature sets. Multi-source feature spectrum refers to the fused feature matrix, including spatiotemporal related features, used for target recognition, and is generated based on feature cross-fusion.

[0155] In this embodiment, the contour feature set, hotspot feature set, node feature set, and icing feature set are first aligned according to spatial coordinates. Secondly, feature cross-fusion is performed through feature concatenation and attention weighting to enhance the correlation of multi-dimensional features. Finally, a multi-source feature spectrum containing spatiotemporal correlation information is generated.

[0156] Step 126: Perform type matching based on multi-source feature spectrum to generate the type of potential attack target.

[0157] Type matching refers to the mapping process between features and target types, including similarity calculation and template comparison, which is used to determine the type of infringement and is obtained based on multi-source feature spectrum analysis.

[0158] In this embodiment, a feature template library is first constructed for four types of targets: foreign objects, external damage, intrusion, and icing. Next, the cosine similarity between the multi-source feature spectrum and the templates is calculated, and the type with the highest matching degree is selected. Finally, type labels for potential threats are generated.

[0159] Here is a specific example: First, a drone was used to capture visible light and infrared images of towers 3 to 5 of the 220kV line, simultaneously receiving dielectric data of the insulators on tower 3 and information on the ice thickness on tower 4. Next, image recognition technology was used to extract the geometric contour features of the kite suspended on tower 3. The hot spot feature, formed by the heated area of ​​the kite string in the infrared image, was detected. Combined with the abnormal dielectric value of the insulators on tower 3, an insulation defect feature was generated. Simultaneously, the measured ice thickness of 18 mm on tower 4, exceeding the safety threshold, generated an ice disaster feature. Then, the four types of features were aligned and fused according to geographical coordinates: the location of tower 3 was associated with the kite contour and hot spot features; the location of the same tower was associated with the insulation defect feature; and the location of tower 4 was associated with the ice feature, forming a spatially bound multi-source feature spectrum. Finally, by comparing with a feature template library, the feature combination of contour anomaly, heat source signal, and insulation anomaly on tower 3 was identified. A "foreign object attachment" type feature template was matched, triggering a foreign object alarm signal for tower 3. The alarm signal will serve as the input condition for subsequent threat value calculation. The contour features are derived from edge detection of visible light images, the hot spot features are derived from temperature segmentation of infrared images, the insulation features are calculated based on the rate of change of dielectric constant, the icing features are generated according to the measured ice thickness, and the feature fusion strictly follows the spatial coordinate mapping relationship of the equipment.

[0160] By executing steps 121 to 126, this embodiment of the application achieves comprehensive perception and accurate classification of transmission line anomalies by integrating multi-dimensional features of visible light, infrared, dielectric and icing, thereby improving the accuracy of identifying foreign object intrusion, equipment damage and natural disasters, and enhancing the power grid's safety protection capabilities.

[0161] In one possible embodiment, S13, when there are multiple potential targets, calculate the basic threat value of each potential target, including:

[0162] Step 131: Quantify the spatial threat of each potential target by analyzing its spatial location parameters and generate corresponding spatial threat factors.

[0163] Spatial location parameters refer to the target's positioning data in a three-dimensional coordinate system, including latitude and longitude, altitude, and distance to critical equipment. These are used for spatial threat analysis and are obtained based on measurements from a positioning system. Spatial threat quantification refers to the process of converting location data into threat values, including distance reciprocal calculation and regional weight allocation. This is used to assess locational hazard and is calculated based on the spatial location parameters. Spatial threat factors are standardized numerical values ​​of locational threats, including normalized hazard indices. These are used for threat fusion and are generated based on spatial threat quantification.

[0164] In this embodiment, the three-dimensional coordinates of each potential target and its distance from the power transmission equipment are first obtained. Next, spatial threat quantification is performed by multiplying the inverse of the distance between the target and the conductor by a location-criticality weighting coefficient. Then, the quantified values ​​are normalized based on the distribution of equipment functional areas. Finally, a spatial threat factor dataset characterizing the degree of locational danger is generated.

[0165] Step 132: Extract the dynamic behavioral characteristics of each potential target and generate the corresponding dynamic threat factor.

[0166] Among them, dynamic behavioral characteristics refer to the time-varying attributes of the target's motion state, including velocity vector, acceleration, and trajectory curvature, used to capture abnormal behavior and extracted based on time-series monitoring data. Dynamic threat factors refer to the quantitative indicators of abnormal behavior, including behavioral deviation and speed hazard coefficient, used for real-time threat assessment and calculated based on dynamic behavioral characteristics.

[0167] In this embodiment, the target's movement speed and trajectory are first extracted through time-series image analysis. Next, a comprehensive index of dynamic behavioral characteristics, calculated by multiplying the trajectory offset angle by the velocity value, is determined. Then, abnormal behavior levels are matched against a behavior pattern library. Finally, a set of dynamic threat factors reflecting real-time threat changes is generated.

[0168] Step 133: For potential targets such as foreign objects, excessive ice accumulation, and personnel intrusion on power transmission lines, the spatial threat factor and dynamic threat factor are fused to generate the corresponding basic threat coefficient.

[0169] The basic threat coefficient refers to the target's basic risk assessment value, which includes the results of multi-dimensional threat fusion and is used for subsequent transmission analysis. It is generated based on the fusion of threat factors.

[0170] In this embodiment, potential targets categorized as foreign objects, icing, and intrusion are first screened. Next, the spatial threat factor is multiplied by the dynamic threat factor for feature fusion. Then, a type weighting coefficient is introduced to adjust the product result. Finally, a basic threat coefficient list for this type of target is generated.

[0171] Step 134: For potential targets of external equipment damage, perform attribute threat calculation on the material property parameters of the potential targets to generate attribute threat factors. Then, fuse the spatial threat factor, dynamic threat factor, and attribute threat factor to generate the corresponding basic threat coefficient.

[0172] Material property parameters refer to the physicochemical characteristics of the equipment material, including hardness, brittleness, and corrosion resistance levels. These are used for external damage risk assessment and are derived from material testing reports. Property threat calculation refers to the process of deriving failure risk based on material properties, including vulnerability models and corrosion rate formulas. This quantifies the threat of material defects and is calculated based on material property parameters. Property threat factors are standardized values ​​of material failure risk, including structural defect indices and aging coefficients. These are used for specific external damage assessments of equipment and are generated based on property threat calculations.

[0173] In this embodiment, firstly, material hardness and aging parameters are extracted for targets with external equipment failure. Secondly, an attribute threat factor, calculated by multiplying the corrosion rate by the structural vulnerability, is calculated using a material failure model. Subsequently, the spatial threat factor, dynamic threat factor, and attribute threat factor are weighted and summed. Finally, a dedicated basic threat coefficient for the target of external equipment failure is generated.

[0174] Here is a specific example: First, visible light and infrared images of towers 3 to 5 of the 220kV line were collected using a drone, simultaneously acquiring insulation performance data for tower 3 and ice thickness information for tower 4. Threat quantification was performed on the kite suspended from tower 3: the distance between the kite and the conductor was measured at 0.5 meters, and the spatial threat factor was calculated as 20 using the formula of multiplying the reciprocal of the distance by 100; the 0.2-meter swing amplitude of the kite in the image sequence was analyzed, and a dynamic threat factor of 10 was generated by multiplying the amplitude value by 50; the arithmetic mean of the two factors was taken to obtain a base threat value of 15. Simultaneously, the ice hazard of tower 4 was assessed: based on the measured ice thickness of 18 mm, exceeding the 15 mm safety threshold, and combined with the importance of the equipment location, an ice hazard point threat value of 12 was generated.

[0175] For the adjacent No. 5 tower transformer equipment, an additional structural risk assessment was conducted: Material parameters of the transformer's cast iron casing were extracted. Based on the material aging coefficient of 0.8 and crack depth of 2 mm, the attribute threat factor was calculated as 0.8 × 2 × 10 = 16 using the material vulnerability formula. Combined with the dynamic threat factor of 6 generated from its vibration displacement of 0.3 meters, and the spatial threat factor of 8 from 1.2 meters away from the construction area, a weighted fusion formula was used to generate a foundation external damage threat coefficient of 10.2. Finally, a comprehensive threat sequence covering environmental foreign objects and equipment structural risks was formed: Kite threat value 15 for No. 3, Ice disaster threat value 12 for No. 4, and equipment external damage threat value 10.2 for No. 5.

[0176] By executing steps 131 to 134, this embodiment of the application achieves accurate quantification of environmental threats such as foreign object icing and structural threats to equipment damage by differentially integrating three-dimensional threat factors of space, dynamics and materials, thereby improving the comprehensiveness of power grid risk identification and the scientific nature of assessment.

[0177] Figure 2This application provides a schematic diagram of the structure of a potential intrusion target identification and active defense system for power equipment, as shown in the embodiments of this application. Figure 2 As shown, the system includes:

[0178] The acquisition module 21 is used to acquire visible light imaging, infrared light imaging, dielectric constant and icing thickness of the target area. The target area includes the air area where the transmission line is located and the ground area below the transmission line.

[0179] The analysis module 22 is used to analyze the type of potential targets based on the visible light image, infrared light image, dielectric constant and ice thickness. The types are foreign objects on the power transmission line, excessive ice thickness, external equipment damage and personnel intrusion.

[0180] The calculation module 23 is used to calculate the basic threat value of each potential target when there are multiple potential targets.

[0181] The combination module 24 is used to determine the target threat value of each potential target based on the basic threat value and the degree of influence between each potential target, and to combine the target threat values ​​of the potential targets into a threat value sequence.

[0182] The generation module 25 is used to generate a dynamically reconstructed defense chain based on the threat value sequence and the preset defense means mapping rules, and to carry out corresponding active defense against different potential targets based on the defense chain.

[0183] Figure 2 The aforementioned power equipment potential attack target identification and active defense system can perform... Figure 1 The implementation principle and technical effects of the method for identifying and actively defending against potential threats to power equipment as described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the power equipment potential threat identification and active defense system in the above embodiments are performed have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0184] In one possible design, Figure 2 The power equipment potential hazard identification and active defense system of the embodiment shown can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32.

[0185] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0186] The processing component 32 is used to: acquire visible light imaging, infrared imaging, dielectric constant, and icing thickness of the target area, which includes the air area where the transmission line is located and the ground area below the transmission line; based on the visible light imaging, infrared imaging, dielectric constant, and icing thickness, analyze the type of potential intrusion targets, which are foreign objects on the transmission line, excessive icing, external equipment damage, and personnel intrusion; when there are multiple potential intrusion targets, calculate the basic threat value of each potential intrusion target; based on the basic threat value and the degree of influence between each potential intrusion target, determine the target threat value of each potential intrusion target, and combine the target threat values ​​of the potential intrusion targets into a threat value sequence; based on the threat value sequence and combined with preset defense means mapping rules, generate a dynamically reconstructed defense chain, and perform corresponding active defense against different potential intrusion targets according to the defense chain.

[0187] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0188] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Random Access Memory (RAM), Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0189] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0190] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0191] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0192] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0193] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a method for identifying potential threats to power equipment and proactively defending against them.

[0194] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0195] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0196] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0197] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying potential threats to power equipment and proactively defending against them, characterized in that, include: The visible light image, infrared image, dielectric constant, and icing thickness of the target area are obtained. The target area includes the air area where the transmission line is located and the ground area below the transmission line. Based on the visible light imaging image, infrared light imaging image, dielectric constant, and ice thickness, the types of potential targets of attack are analyzed, namely foreign objects on the transmission line, excessive ice thickness, external equipment damage, and personnel intrusion. When there are multiple potential targets, calculate the base threat value for each potential target; Based on the basic threat value and the degree of influence between each potential target, the target threat value of each potential target is determined, and the target threat values ​​of the potential targets are combined into a threat value sequence. Based on the threat value sequence and the preset defense method mapping rules, a dynamically reconstructed defense chain is generated, and corresponding active defense is carried out against different potential targets based on the defense chain. When there are multiple potential targets, the basic threat value of each potential target is calculated, including: Spatial threat quantification is performed on the spatial location parameters of each potential target to generate a corresponding spatial threat factor; Extract the dynamic behavioral characteristics of each potential target and generate corresponding dynamic threat factors; For potential targets such as foreign objects, excessive ice accumulation, and personnel intrusion on power transmission lines, the spatial threat factors and dynamic threat factors are fused to generate corresponding basic threat coefficients. For potential targets of external damage to equipment, the material property parameters of the potential targets are used to calculate the attribute threat, generate attribute threat factors, and fuse the spatial threat factors, dynamic threat factors and attribute threat factors to generate the corresponding basic threat coefficient.

2. The method according to claim 1, characterized in that, The step of generating a dynamically reconstructed defense chain based on the threat value sequence and a preset defense measure mapping rule includes: Based on the threat value sequence, determine the threat value range to which the target threat value of each potential intrusion target belongs, and generate a set of defense units based on the threat value range to which the target threat value of each potential intrusion target belongs and the corresponding defense resource type; Perform a topology constraint assignment operation on the set of defense units to generate an initial defense chain; The initial defense chain is subjected to threat chain effect calculation between adjacent devices and defense unit priority reordering operation to generate a reconstructed defense chain.

3. The method according to claim 2, characterized in that, The step of performing threat chain effect calculations between adjacent devices and prioritizing defense units on the initial defense chain to generate a reconstructed defense chain includes: Based on the physical connection relationship of the defense unit associated devices in the initial defense chain, the threat propagation path is calculated and a threat propagation path is generated. The threat intensity of the aforementioned threat propagation paths is accumulated to generate a cumulative threat intensity; Based on the accumulated threat intensity, the defense units in the initial defense chain are prioritized and reordered to generate a reconstructed defense chain.

4. The method according to claim 1, characterized in that, The step of determining the target threat value of each potential target based on the basic threat value and the degree of influence between each potential target, and combining the target threat values ​​of the potential targets into a threat value sequence, includes: The influence between the potential targets is analyzed to generate a set of threat transmission paths. Calculate the path weights of the threat propagation path set to generate a path weight set; Based on the path weight set, the basic threat value of each potential target is corrected to generate a target threat value. The target threat values ​​of all the potential targets are sorted by device topology location to generate a threat value sequence.

5. The method according to claim 4, characterized in that, The analysis of the transmission path of the impact between the potential targets generates a set of threat transmission paths, including: Quantify the connection strength of the connection topology among the potential targets of attack to form a set of connection strengths; By using failure modes, the set of connection strengths and the degree of impact are correlated to generate a failure correlation set; Based on the failure association set, a propagation path expansion operation is performed to generate an initial propagation path set; The initial transmission path set is modified by environmental interference to generate the threat transmission path set.

6. The method according to claim 1, characterized in that, The method of analyzing the type of potential target based on the visible light image, infrared image, dielectric constant, and icing thickness includes: Extract the shape and contour features from the visible light image to obtain a set of contour features that characterize whether there are foreign objects, whether the equipment is damaged, and whether personnel have intruded on the transmission line. Perform a temperature anomaly region segmentation operation on the infrared light image to generate a set of hot spot features that characterize whether there are foreign objects or personnel intrusion on the transmission line; Based on the dielectric constant, determine the set of node features that characterize the insulation degradation region where there is external equipment damage; Perform ice damage risk classification on the ice thickness to generate a set of ice features that characterize whether there is an ice risk. The contour feature set, the hot spot feature set, the node feature set, and the icing feature set are cross-fused to generate a multi-source feature spectrum; Type matching is performed based on the multi-source feature spectrum to generate the type of potential attack targets.

7. A system for identifying and actively defending against potential threats to power equipment, characterized in that, A method for identifying and actively defending against potential threats to power equipment as described in any one of claims 1 to 6, comprising: The acquisition module is used to acquire visible light imaging, infrared light imaging, dielectric constant and icing thickness of the target area, wherein the target area includes the air area where the transmission line is located and the ground area below the transmission line; The analysis module is used to analyze the type of potential intrusion target based on the visible light imaging image, infrared light imaging image, dielectric constant and ice thickness, wherein the type is foreign object on the transmission line, excessive ice thickness, external equipment damage and personnel intrusion; The calculation module is used to calculate the base threat value of each potential target when there are multiple potential targets. The combination module is used to determine the target threat value of each potential target based on the basic threat value and the degree of influence between each potential target, and to combine the target threat values ​​of the potential targets into a threat value sequence. The generation module is used to generate a dynamically reconstructed defense chain based on the threat value sequence and a preset defense method mapping rule, and to perform corresponding active defense against different potential targets based on the defense chain.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the method for identifying and actively defending against potential threats to power equipment as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements a method for identifying and actively defending against potential threats to power equipment as described in any one of claims 1 to 6.

Images