A method, device, equipment and medium for determining a risk blind area of power grid training

Abstract

A method, device, equipment and medium for determining a risk blind area of power grid training are disclosed. The method comprises: obtaining eye gaze data of a user in a virtual electric field scene; determining an initial visual focus area of the user according to the eye gaze data, and generating a hidden defect area corresponding to the virtual power grid scene in the initial visual focus area; obtaining equipment state parameters, material parameters and operation load data of the user in the hidden defect area; performing damage detection on power grid facilities in the hidden defect area to obtain an abnormal detection result of the hidden defect area; when the abnormal detection result is damage, obtaining a target visual focus area of the user at an accident triggering moment, and an equipment damage area of the hidden defect area; and determining a corresponding risk blind area of the user according to a difference between the target visual focus area and the equipment damage area. The risk blind area formed by the trainee not discovering potential risks can be identified.

Description

Technical Field

[0001] This invention relates to the field of virtual reality training technology, and in particular to a method, apparatus, equipment and medium for determining risk blind spots in power grid training. Background Technology

[0002] With the continuous expansion of power system scale and the increasing complexity of power grid equipment, power grid operation, maintenance, and accident handling personnel need to possess high levels of hazard identification and on-site operational skills. Traditional training methods based on real equipment suffer from high training costs, significant risks, and difficulty in replicating complex accident scenarios. Therefore, virtual power grid training scenarios built using virtual reality technology are increasingly being applied in power grid personnel training. Virtual reality technology can digitally simulate the structure of power grid facilities, operating environment, and accident processes, enabling trainees to conduct equipment inspections, troubleshooting, and emergency operations in a virtual environment, thereby improving the safety and effectiveness of the training process.

[0003] In existing technologies, virtual reality-based power grid training typically involves pre-constructing a virtual power grid scenario that includes faulty equipment and accident processes. Based on a pre-defined accident script, equipment defects, abnormal states, or accident outcomes are displayed at fixed times or locations. Trainees interact with the virtual scenario using virtual reality devices, making judgments and taking actions based on the fault information displayed by the system. The training system then evaluates the trainees' fault-handling capabilities based on their operational results.

[0004] However, existing technologies typically evaluate training based solely on the trainees' final operational results. They cannot further determine whether trainees paid attention to areas where equipment actually exists, nor can they identify blind spots caused by trainees failing to detect potential risks. This affects the accuracy of training evaluations and the effectiveness of improving hazard identification capabilities. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for determining risk blind spots in power grid training. The embodiments of this invention can identify risk blind spots formed by trainees failing to discover potential risks.

[0006] In a first aspect, embodiments of the present invention provide a method for determining risk blind spots in power grid training, the method comprising:

[0007] Acquire user eye gaze data in a virtual electric field scene;

[0008] Based on eye gaze data, the user's initial visual attention area is determined, and a hidden defect area corresponding to the virtual power grid scene is generated in the initial visual attention area.

[0009] Acquire equipment status parameters, material parameters, and user operating load data within the area of ​​hidden defects;

[0010] Damage detection is performed on power grid facilities in areas with latent defects based on equipment status parameters, material parameters, and operating load data, and abnormal detection results are obtained for these areas.

[0011] When the anomaly detection result indicates that damage has occurred, the user's target visual attention area at the time of the incident is obtained, as well as the equipment damage area of ​​the hidden defect area.

[0012] Based on the difference between the target visual focus area and the equipment damage area, the corresponding risk blind spot for the user is determined.

[0013] Secondly, embodiments of the present invention also provide a device for determining risk blind spots in power grid training, the device comprising:

[0014] The eye gaze data acquisition module is used to acquire the user's eye gaze data in the virtual electric field scene;

[0015] The initial visual attention area determination module is used to determine the user's initial visual attention area based on eye gaze data, and generate the hidden defect area corresponding to the virtual power grid scene in the initial visual attention area.

[0016] The latent defect area data acquisition module is used to acquire equipment status parameters, material parameters, and user operating load data within the latent defect area.

[0017] The anomaly detection result determination module is used to perform damage detection on the power grid facilities in the latent defect area based on equipment status parameters, material parameters and operating load data, and obtain the anomaly detection results of the latent defect area.

[0018] The target visual attention area acquisition module is used to acquire the user's target visual attention area and the equipment damage area of ​​the hidden defect area at the time of the accident trigger when the anomaly detection result is that damage has occurred.

[0019] The risk blind spot determination module is used to determine the user's corresponding risk blind spot based on the difference between the target visual attention area and the equipment damage area.

[0020] Thirdly, embodiments of the present invention also provide a risk blind spot determination device for power grid training, the risk blind spot determination device for power grid training comprising:

[0021] At least one processor; and

[0022] A memory that is communicatively connected to at least one processor; wherein,

[0023] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the risk blind spot determination method for power grid training according to any embodiment of the present invention.

[0024] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute the risk blind spot determination method for power grid training according to any embodiment of the present invention.

[0025] Fifthly, embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements a method for determining risk blind spots in power grid training according to any embodiment of the present invention.

[0026] The technical solution of this invention acquires user eye gaze data in a virtual power grid scenario, determines the user's visual attention area based on the eye gaze data, generates a corresponding latent defect area in the user's visual attention area, and then performs damage detection on the latent defect area by combining equipment status parameters, material parameters, and user operation load data. This enables the defect setting in the virtual training process to be combined with the user's actual attention behavior. When damage occurs, by acquiring the user's target visual attention area and the equipment damage area, and determining the user's corresponding risk blind spot based on the difference between the two, it can identify areas that the user did not pay attention to but actually suffered damage, achieving accurate determination of the user's risk blind spot and improving the accuracy of risk identification analysis in the virtual power grid training process. This solves the technical problem of existing technologies that cannot identify risk blind spots formed by trainees not discovering potential risks, thus enabling the identification of risk blind spots formed by trainees not discovering potential risks.

[0027] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating a method for determining risk blind spots in power grid training, provided as an embodiment of the present invention;

[0030] Figure 2 A flowchart illustrating a method for determining risk blind spots in power grid training, provided as an embodiment of the present invention;

[0031] Figure 3 A schematic diagram of a risk blind spot determination device for power grid training provided in an embodiment of the present invention;

[0032] Figure 4 This is a structural schematic diagram of a risk blind spot determination device for power grid training provided in an embodiment of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] In the technical solutions of this invention, the acquisition, storage, and application of eye gaze data, device status parameters, material parameters, and operational load data all comply with relevant laws and regulations and do not violate public order and good morals.

[0036] Figure 1 This is a flowchart illustrating a method for determining risk blind spots in power grid training, provided by an embodiment of the present invention. This embodiment is applicable to situations where risk blind spots are determined for users during power grid training using virtual reality devices. The method can be executed by a power grid training risk blind spot determination device, which can be implemented in hardware and / or software.

[0037] See Figure 1 The method for identifying risk blind spots in power grid training, as shown, includes:

[0038] S101. Obtain the user's eye gaze data in the virtual electric field scene.

[0039] In this context, a virtual power grid scenario refers to a digital 3D scene constructed using computer graphics simulation technology to simulate the operating environment of real power grid equipment and the training process. This virtual power grid scenario typically includes 3D models of power grid facilities such as poles, transmission lines, transformers, distribution equipment, tower foundations, and guy wires, as well as scene data describing the spatial location, connection relationships, and operating status of the equipment. For example, a distribution line scenario including poles, guy wires, and foundation structures can be constructed in a virtual environment. After entering this scenario wearing virtual reality equipment, users can observe the equipment status and perform corresponding operations.

[0040] Eye gaze data refers to a dataset describing a user's eye movement and visual observation behavior, reflecting changes in the user's gaze within a virtual power grid scenario. This data typically includes information such as the coordinates of the user's gaze point, gaze direction, gaze timestamp, and gaze duration.

[0041] S102. Based on eye gaze data, determine the user's initial visual attention area, and generate the hidden defect area corresponding to the virtual power grid scene in the initial visual attention area.

[0042] The initial visual attention area (IGA) refers to the virtual power grid scene area that the user focused on before the accident, determined based on user eye gaze data. This area characterizes the location or spatial range of equipment that the user actively focused on during training. The IGA typically corresponds to one or more grid regions in the 3D model of the virtual power grid scene, and can consist of multiple spatially continuous or discretely distributed regions. This area corresponds to scene locations with high gaze intensity or high gaze confidence factors.

[0043] In this context, a latent defect area refers to a defect region generated in a virtual power grid scenario that is not initially visible to the user but can affect the structural state of the equipment. This area is used to simulate internal damage or potential risks in real power grid facilities that are difficult to detect visually. For example, when a user observes the middle area of ​​a utility pole, an internal crack defect can be generated in that area, and the structural damage caused by crack propagation can be simulated during subsequent operations.

[0044] S103. Obtain equipment status parameters, material parameters, and user operating load data within the area of ​​hidden defects.

[0045] Among these, equipment status parameters refer to data parameters that describe the current status and defects of virtual power grid facilities. These parameters reflect status information such as whether the equipment has defects, the location of the defects, and the degree of defects. For example, for a pole with internal cracks, the location and severity of the cracks can be determined through equipment status parameters, thereby calculating the development of the cracks under external loads.

[0046] Material parameters refer to data parameters describing the physical properties of the materials that make up power grid facilities. These parameters characterize the ability of equipment materials to resist external loads and undergo structural deformation. For example, when the same operating load is applied to equipment made of different materials, different structural responses can be calculated based on the material parameters.

[0047] Operational load data refers to the external forces exerted on power grid facilities by users when performing operations in a virtual power grid scenario. Operational load data is used to describe the impact of user behavior on equipment structure.

[0048] S104. Based on equipment status parameters, material parameters, and operating load data, conduct damage detection on the power grid facilities in the latent defect area to obtain the abnormal detection results of the latent defect area.

[0049] Damage detection can refer to the process of analyzing and judging whether a virtual power grid facility has suffered structural damage in its current state, based on equipment status parameters, material parameters, and operating load data.

[0050] Anomaly detection results can refer to the judgment results obtained from the damage detection process, indicating whether the power grid facilities corresponding to the latent defect area have been damaged. Anomaly detection results can include two states: damage occurred and no damage occurred, and can further include information such as the degree of damage and the location of damage.

[0051] S105. When the abnormal detection result indicates that damage has occurred, obtain the user's target visual attention area at the time of the accident triggering, as well as the equipment damage area of ​​the hidden defect area.

[0052] The target visual attention area can refer to the virtual power grid scene area that the user is actually observing at the moment the accident is triggered. The target visual attention area reflects the user's true visual focus location at the instant the accident occurs.

[0053] In this context, the equipment damage area refers to the spatial region within a virtual power grid scenario where abnormal changes in the structural performance of power grid facilities occur due to the development of latent defects and the effects of user operational loads. The equipment damage area represents the actual location and extent of structural damage to the power grid facilities.

[0054] S106. Based on the difference between the target visual attention area and the equipment damage area, determine the user's corresponding risk blind spot.

[0055] In this context, a risk blind spot refers to an area where equipment is actually damaged during virtual grid training, but the user fails to pay effective visual attention to it. A risk blind spot indicates a location missed by the user during risk identification. For example, if an internal crack in equipment expands to a dangerous state, but the user's line of sight did not cover the crack location before the accident, then that crack area corresponds to a risk blind spot.

[0056] As can be seen, in this embodiment, by acquiring the user's eye gaze data in the virtual power grid scenario and determining the user's visual attention area based on the eye gaze data, a corresponding latent defect area is generated in the user's visual attention area. Then, by combining equipment status parameters, material parameters, and user operational load data, damage detection is performed on the latent defect area. This allows the defect settings in the virtual training process to be combined with the user's actual attention behavior. When damage occurs, by acquiring the user's target visual attention area and the equipment damage area, and determining the user's corresponding risk blind spot based on the difference between the two, it is possible to identify areas that the user did not pay attention to but which actually suffered damage, thus achieving accurate determination of the user's risk blind spot and improving the accuracy of risk identification analysis during virtual power grid training. This solves the technical problem of existing technologies being unable to identify risk blind spots formed by trainees failing to detect potential risks, thereby enabling the identification of risk blind spots formed by trainees failing to detect potential risks.

[0057] In an optional embodiment, Figure 2 The flowchart of a method for determining risk blind spots in power grid training provided by an embodiment of the present invention refines the process of "determining the risk blind spot corresponding to the user based on the difference between the target visual attention area and the equipment damage area" into "screening based on the damage value corresponding to the equipment damage area and a preset failure threshold to obtain an initial physical damage grid group; performing connectivity analysis on the spatial adjacency relationship between at least one grid cell in the initial physical damage grid group to obtain at least one damage connected region; and performing spatial overlap difference processing on the damage connected region and the target visual attention area to obtain the user's risk blind spot," thereby improving the operation of determining risk blind spots in power grid training.

[0058] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments.

[0059] See Figure 2 The method for identifying risk blind spots in power grid training, as shown, includes:

[0060] S201. Obtain the user's eye gaze data in the virtual electric field scene.

[0061] S202. Based on eye gaze data, determine the user's initial visual attention area, and generate the hidden defect area corresponding to the virtual power grid scene in the initial visual attention area.

[0062] S203. Obtain equipment status parameters, material parameters, and user operating load data within the area of ​​hidden defects.

[0063] S204. Based on equipment status parameters, material parameters, and operating load data, damage detection is performed on the power grid facilities in the latent defect area to obtain the abnormal detection results of the latent defect area.

[0064] S205. When the abnormal detection result indicates that damage has occurred, obtain the user's target visual attention area at the time of the accident triggering, as well as the equipment damage area of ​​the hidden defect area.

[0065] S206. Based on the damage values ​​corresponding to the damaged areas of the equipment and the preset failure threshold, filter to obtain the initial physical damage grid group.

[0066] The damage value refers to a numerical parameter used to quantify the degree of structural damage at various spatial locations within a virtual power grid facility. The damage value reflects the severity of damage to different areas of the equipment.

[0067] The preset failure threshold can be a pre-defined critical value used to determine whether a local area of ​​the power grid facility has reached a dangerous damage state. The preset failure threshold is used to distinguish between normal operating areas and damaged areas that may lead to accidents.

[0068] The initial physical damage grid set can refer to a dataset containing high-risk damage grid cells obtained after filtering based on damage values ​​and preset failure thresholds. By filtering out minor damage areas using damage thresholds, only high-risk grid cells that could lead to accidents are retained, thereby improving the accuracy of subsequent risk blind spot identification.

[0069] S207. Perform connectivity analysis on the spatial adjacency relationship between at least one grid cell in the initial physical damage grid group to obtain at least one damaged connected region.

[0070] Spatial adjacency refers to the spatial contact or interconnection between different mesh elements in a virtual power grid 3D model. Spatial adjacency describes the continuity of damaged mesh elements in 3D space. For example, multiple adjacent mesh elements corresponding to continuous cracks on a power pole can be identified as the same damaged region based on spatial adjacency.

[0071] Connectivity analysis can refer to the aggregation and classification of grid cells with connectivity based on spatial adjacency to determine the processing method for multiple independent regions.

[0072] In this context, the damage connectivity region refers to a complete equipment damage area composed of multiple spatially continuous damage mesh cells, obtained after connectivity analysis. The purpose of the damage connectivity region is to avoid errors caused by judging based on a single damage point, enabling risk blind spot identification to be performed on the complete damage area.

[0073] S208. Perform spatial overlap difference processing on the damaged connected region and the target visual interest region to obtain the user's risk blind zone.

[0074] Spatial overlap differential processing refers to the process of calculating the spatial coverage relationship between the damaged area of ​​the device and the area of ​​visual focus of the user, and determining the risk blind zone based on the difference between the two. Spatial overlap differential processing is used to determine whether the user has noticed the actual location of the damage. By comparing the degree of consistency between the physical damage area and the user's perceived area of ​​focus, it identifies the mismatch between physical risk and user perception. For example, when the area of ​​the damaged connected region is large, but has almost no overlap with the user's perceived area, it indicates that the user has not noticed the risk area.

[0075] As can be seen, in this embodiment, by filtering based on the damage value corresponding to the damaged area of ​​the equipment and the preset failure threshold, physical damage grid units that reach the risk level can be identified from the damaged area of ​​the equipment, reducing the impact of minor damage or invalid damage data on risk assessment; furthermore, based on the spatial adjacency relationship between grid units, connected component analysis is performed to aggregate the scattered damage grid units into complete damage connected regions, which can more accurately characterize the spatial distribution of the actual damaged area of ​​the equipment; by performing spatial overlap difference processing on the damage connected regions and the target visual attention area, inconsistent areas between the actual damaged area of ​​the equipment and the user's visual attention area can be identified, thereby determining risk blind spots that the user has not effectively paid attention to but where there is a risk of equipment damage, improving the accuracy of risk blind spot determination.

[0076] In some embodiments, determining the user's initial visual attention area based on eye gaze data includes:

[0077] Based on eye gaze data, determine at least one gaze point corresponding to the user;

[0078] Divide the virtual power grid scenario into at least one candidate region;

[0079] Based on the location coordinates of the fixation point, determine the number of fixations and fixation duration for the candidate region;

[0080] The fixation intensity of the candidate region is obtained by weighting the number of fixations and the fixation duration of the candidate region.

[0081] Based on at least one gaze intensity, determine the user's initial visual attention area from at least one candidate region.

[0082] The fixation point can refer to the projection position of the user's eye on the virtual power grid scene, which is used to represent the specific spatial location of the user's current visual observation.

[0083] Among them, the number of fixations can refer to the cumulative number of times a user's gaze falls into a certain candidate area, which is used to indicate the degree of repeated observation of the area by the user.

[0084] Among them, gaze duration can refer to the length of time a user's gaze remains on a candidate area, which is used to indicate the degree of duration of the user's observation of that area.

[0085] Fixation intensity refers to an evaluation parameter that comprehensively reflects the degree of attention a user pays to a candidate area. Fixation intensity is used to filter the areas of actual user focus from multiple candidate areas. For example, by comprehensively calculating the number of fixations and the duration of fixation on a certain area, it can be determined whether that area belongs to the user's key observation area.

[0086] As can be seen, in this embodiment, by determining at least one gaze point corresponding to the user based on eye gaze data, and combining it with candidate regions obtained from the virtual power grid scene, the user's visual attention behavior can be associated with specific regions in the virtual scene. Furthermore, by statistically analyzing the gaze frequency and duration of the candidate regions based on the position coordinates of the gaze point, and by weighting the gaze frequency and duration to obtain the gaze intensity, the user's attention to different regions can be comprehensively reflected, avoiding inaccurate visual attention judgments based solely on a single gaze parameter. Determining the user's initial visual attention region based on the gaze intensity can more accurately identify the user's main attention region in the virtual power grid scene, providing a data foundation for subsequently generating latent defects in the corresponding regions and conducting risk blind spot analysis.

[0087] In some embodiments, determining the user's initial visual attention region from at least one candidate region based on at least one gaze intensity includes:

[0088] Calculate the fixation confidence factor for candidate regions;

[0089] Based on the gaze confidence factor of at least one candidate region and a preset attention threshold, at least one candidate region is filtered to determine the user's initial visual attention region.

[0090] The fixation confidence factor is an evaluation parameter used to represent the credibility of a user's attention to a candidate region. It is typically calculated based on fixation intensity, gaze stability, and observation duration, and its value ranges from 0 to 1; a higher value indicates a higher credibility of the user's attention to that region.

[0091] The preset attention threshold refers to a pre-defined critical parameter used to determine whether a user has formed effective visual attention to a certain area in the virtual power grid scene. The preset attention threshold is used to filter candidate areas to determine the initial visual attention area that the user will actually focus on. The gaze confidence factor corresponding to each candidate area is compared with this threshold; when the gaze confidence factor reaches the preset attention threshold, the user is considered to have formed effective attention to that area.

[0092] As can be seen, in this embodiment, by calculating the gaze confidence factor of the candidate region, the visual attention of the user in different candidate regions can be quantitatively represented, making the attention levels between different regions comparable. Furthermore, by combining the preset attention threshold to screen the candidate regions, regions with low user attention can be filtered out, and regions that meet the attention conditions can be retained as the initial visual attention regions, thereby improving the accuracy of determining the user's visual attention regions and providing reliable data for the subsequent generation of latent defect regions based on the user's attention regions.

[0093] In some embodiments, damage detection is performed on power grid facilities in areas with latent defects based on equipment status parameters, material parameters, and operating load data to obtain anomaly detection results for the latent defect areas, including:

[0094] Based on the material parameters, determine the material mechanical parameters of the power grid facilities corresponding to the hidden defect areas;

[0095] Based on the equipment status parameters, determine the defect parameters corresponding to the latent defect areas;

[0096] Based on material mechanical parameters, defect parameters, and operating load data, physical evolution calculations are performed on the power grid facilities corresponding to the hidden defect areas to obtain stress distribution data, strain distribution data, and displacement data corresponding to the power grid facilities.

[0097] Based on stress distribution data, strain distribution data, and displacement data, the actual structural damage field corresponding to the hidden defect region is generated.

[0098] By comparing the damage degree parameters in the actual structural damage field with the preset accident trigger threshold, the abnormal detection results corresponding to the hidden defect area are obtained.

[0099] Material mechanical parameters refer to parameters used to describe the mechanical properties of materials used in power grid facilities, reflecting the load-bearing capacity and deformation characteristics of the materials under external forces. Material mechanical parameters include, but are not limited to, compressive strength, tensile strength, modulus of elasticity, yield strength, and shear strength. Different types of power grid facilities correspond to different material mechanical parameters; for example, concrete poles correspond to compressive strength and modulus of elasticity, steel reinforcement corresponds to yield strength and tensile strength, and foundation soil corresponds to compressive strength and shear strength.

[0100] Among them, defect parameters refer to information parameters that describe the defect state within a latent defect area, used to indicate the initial abnormal state of power grid facilities. Defect parameters include information such as defect type, defect location, defect size, defect level, and defect impact range.

[0101] The change in load-bearing capacity refers to the change in the ability of power grid facilities to withstand external loads after the presence of latent defects. This change in load-bearing capacity describes the impact of latent defects on the structural performance of equipment. Latent defects reduce the mechanical properties of localized areas of the equipment, thus requiring adjustments to the actual load-bearing capacity based on the defect status. For example, internal cracks in a power pole can reduce the tensile strength of the cracked area, making the equipment more susceptible to damage under relatively small loads.

[0102] The real structural damage field refers to a data field representing the actual damage state of virtual power grid facilities at various spatial locations during their physical evolution. The real structural damage field is used to reflect the actual damage development of latent defect regions after the application of operational loads.

[0103] The damage severity parameter can be a parameter that quantifies the severity of damage at various locations within a real structural damage field. This parameter is used to determine whether power grid facilities have reached the conditions for triggering an accident.

[0104] Among them, the preset accident trigger threshold can refer to the pre-set critical parameter for determining whether the power grid facilities have entered an accident state.

[0105] As can be seen, in this embodiment, by determining the material mechanical parameters of the power grid facility corresponding to the latent defect area based on material parameters, and combining them with equipment state parameters to determine the corresponding defect parameters, the material properties of the power grid facility itself and the state of latent defects can be incorporated into the damage detection process, improving the completeness of parameter acquisition during damage detection. Furthermore, by performing physical evolution calculations based on material mechanical parameters, defect parameters, and operating load data, the internal stress, strain, and displacement changes of the power grid facility under operating loads can be simulated, obtaining stress distribution data, strain distribution data, and displacement data that reflect the stress state of the facility. Based on the above data, a real structural damage field is generated, and by comparing the damage degree parameters with a preset accident triggering threshold, it can be determined whether the power grid facility corresponding to the latent defect area has reached the accident triggering condition, thereby realizing the detection of the damage state of the latent defect area and improving the accuracy of anomaly detection results.

[0106] In some embodiments, after determining the risk blind spot corresponding to the user, the method further includes:

[0107] Generate corresponding cognitive blind spot warning masks based on risk blind spots;

[0108] Obtain the real-time rendered screen that was locked at the moment the accident was triggered, and unfreeze the frozen graphics rendering thread;

[0109] Based on the cognitive blind spot warning mask and real-time rendering, the accident backtracks and re-enacts the evolution from the moment the accident was triggered to obtain the accident backtracking scenario corresponding to the risk blind spot;

[0110] Adjust the backtracking speed of the accident backtracking scenario based on the backtracking control commands input by the user;

[0111] The cognitive blind spot warning mask is overlaid on the rendering screen corresponding to the accident retrospective scene to show the hidden danger area corresponding to the risk blind spot and the damage evolution process of the hidden danger area.

[0112] Among them, the cognitive blind spot warning mask refers to a data mask used to identify areas of equipment that are not noticed by the user but are actually at risk of damage. The cognitive blind spot warning mask is used to highlight risk areas that the user has overlooked during accident review. After analyzing the differences between the physically damaged areas and the areas the user is visually focused on, corresponding labels are generated and overlaid onto the virtual scene. For example, crack areas that the user did not observe are highlighted, allowing the user to clearly identify the location they have visually missed.

[0113] Real-time rendered footage refers to the visual display generated in real time during virtual power grid training based on the 3D model of the virtual power grid, user operations, and rendering parameters. Real-time rendered footage is used to record the virtual training scenario at the moment an accident is triggered and serves as the basic visual data for accident retrospective reconstruction.

[0114] In this context, a frozen graphics rendering thread refers to a graphics processing flow that pauses real-time rendering updates of the virtual scene when an incident is triggered. The graphics rendering thread includes processes such as model loading, texture processing, lighting calculation, and image output. A frozen state indicates that the aforementioned real-time update process has stopped, keeping the current image in the state it was at the moment the incident was triggered.

[0115] In this context, an accident retrospective scenario refers to a virtual scenario that re-demonstrates the development of equipment damage before the accident, based on the virtual power grid state at the moment the accident was triggered. Accident retrospective scenarios help users understand the causes of accidents and identify undetected risk areas. By retracing time backward from the moment the accident was triggered, the evolution of the accident is gradually reconstructed according to the development of equipment damage, and the corresponding risk blind spots are overlaid and displayed.

[0116] As can be seen, in this embodiment, by generating a corresponding cognitive blind spot warning mask based on the risk blind spot, areas that users have not paid attention to but have the risk of equipment damage can be identified; furthermore, the real-time rendering screen locked at the moment of accident triggering is obtained, and the frozen graphics rendering thread is unfrozen, so that the virtual power grid scene can resume rendering function, providing a visual basis for accident process retrospective; based on the cognitive blind spot warning mask and the real-time rendering screen, the evolution can be retrospectively reproduced from the moment of accident triggering, which can restore the damage development process of the hidden danger area corresponding to the risk blind spot, and the retrospective speed can be adjusted in combination with the retrospective control command input by the user, improving the flexibility of accident process analysis; by superimposing the cognitive blind spot warning mask onto the rendering screen corresponding to the accident retrospective scene, the area that the user has not paid attention to and the damage evolution process of the area can be intuitively displayed, helping the user to identify the risk blind spot caused by insufficient visual attention.

[0117] In some embodiments, generating a latent defect region corresponding to the virtual power grid scene in the initial visual attention region includes:

[0118] Based on the gaze confidence factor, preset disclosure threshold and transition bandwidth parameter corresponding to the initial visual attention area, interval mapping calculation is performed to obtain the visual mixing coefficient matrix corresponding to at least one grid area in the 3D model of the virtual power grid scene.

[0119] Based on at least one visual blending coefficient matrix, an instantiated rendering indirect command buffer is generated. Based on the target mesh region in the instantiated rendering indirect command buffer where the blending coefficient is greater than a preset value, the corresponding micro-damage geometric displacement data and anisotropic crack normal data are matched from the latent defect model database to obtain a high-frequency texture resource package.

[0120] Based on the high-frequency texture resource package, the graphics processing unit performs texture decompression and video memory residency, and establishes a binding relationship between the texture resource identifier corresponding to the high-frequency texture resource package and the grid vertex coordinates in the 3D model of the virtual power grid scene;

[0121] Configure the defect visualization shader according to the binding relationship, and pass the local stress concentration coefficients corresponding to the hidden defect model into the defect visualization shader to obtain the active rendering context;

[0122] Pixel-level fusion rendering is performed based on the active rendering context, visual blending coefficient matrix, and preset rendering optimization instructions to generate hidden defect areas corresponding to the virtual power grid scene.

[0123] The preset disclosure threshold refers to a pre-defined parameter used to determine whether latent defects in a virtual power grid scenario need to be visualized. This threshold controls the generation range of latent defects based on the user's visual attention level. By comparing the gaze confidence factor corresponding to the initial visual attention area with this threshold, the area where the latent defect display effect needs to be generated is determined.

[0124] The instantiation rendering indirect command buffer refers to a data cache area used to store rendering instructions required by the graphics processing unit to perform instantiation rendering operations. The instantiation rendering indirect command buffer is used to improve the real-time rendering efficiency of areas with hidden defects.

[0125] The latent defect model database refers to a dataset used to store various latent defect model data in a virtual power grid facility. This database is used to match corresponding defect data based on the equipment type and defect type of the target grid area.

[0126] Anisotropic crack normal data refers to data used to describe the variation of normals in different directions on the crack surface, thereby achieving a realistic light reflection effect in the crack region. Anisotropic crack normal data is used to improve the visual realism of areas with latent defects.

[0127] High-frequency texture resource packages refer to high-precision texture data sets used to describe the microstructural features of latent defects. These packages include resource information such as micro-damage geometric permutation data, crack texture data, and anisotropic normal data.

[0128] The active rendering context refers to a set of data that describes the current graphics rendering state. It includes information such as the currently used shader configuration, texture resource information, model vertex data, rendering parameters, and graphics processing status, providing a complete runtime environment for subsequent rendering operations.

[0129] The visual blending coefficient matrix refers to a data matrix used to represent the degree of display of latent defects in each grid region of a virtual power grid scene's 3D model. The visual blending coefficient matrix corresponds to the grid regions in the virtual power grid's 3D model, with each matrix element corresponding to a defect display weight for that grid region. Its value typically ranges from 0 to 1. A larger value indicates a higher degree of display of latent defects in that region.

[0130] Among them, preset rendering optimization instructions refer to pre-defined processing instructions used to improve the real-time rendering efficiency of virtual power grid scenes. Preset rendering optimization instructions are used to reduce the computational burden during the real-time rendering process of hidden defects.

[0131] As can be seen, in this embodiment, by performing interval mapping calculations based on the gaze confidence factor corresponding to the initial visual attention area, the preset disclosure threshold, and the transition bandwidth parameter, the visual blending coefficients corresponding to different grid areas can be determined according to the user's visual attention level, thus associating the presentation range of latent defects with the user's visual attention area. Furthermore, by generating an instantiated rendering indirect command buffer based on the visual blending coefficient matrix and matching the corresponding microscopic damage geometric permutation data and anisotropic crack normal data from the latent defect model database, the corresponding defect texture data can be called for different grid areas, improving the accuracy of latent defect model generation. By utilizing graphics... The processing unit decompresses and stores high-frequency texture resource packages in video memory, and establishes a binding relationship between texture resource identifiers and mesh vertex coordinates, which can improve the efficiency of loading and mapping latent defect textures. Furthermore, it configures a defect visualization shader based on the binding relationship and generates an active rendering context by combining local stress concentration coefficients, so that the rendering effect of latent defects can be associated with defect status information. By performing pixel-level fusion rendering based on the active rendering context, visual blending coefficient matrix and preset rendering optimization instructions, it can generate latent defect regions that match the user's visual focus area, providing a basis for subsequent damage detection and risk blind zone determination based on latent defect regions.

[0132] Figure 3 This invention provides a schematic diagram of a risk blind spot determination device for power grid training. This invention is applicable to situations where risk blind spots are determined for training participants using virtual reality equipment. The device can execute a risk blind spot determination method for power grid training and can be implemented in hardware and / or software.

[0133] See Figure 3The power grid training risk blind spot determination device shown includes: an eye gaze data acquisition module 301, an initial visual attention area determination module 302, a latent defect area data acquisition module 303, an anomaly detection result determination module 304, a target visual attention area acquisition module 305, and a risk blind spot determination module 306, wherein...

[0134] The eye gaze data acquisition module 301 is used to acquire the user's eye gaze data in the virtual electric field scene;

[0135] The initial visual attention area determination module 302 is used to determine the user's initial visual attention area based on eye gaze data, and generate a hidden defect area corresponding to the virtual power grid scene in the initial visual attention area.

[0136] The latent defect area data acquisition module 303 is used to acquire equipment status parameters, material parameters, and user operating load data within the latent defect area;

[0137] The anomaly detection result determination module 304 is used to perform damage detection on the power grid facilities in the latent defect area based on equipment status parameters, material parameters and operating load data, and obtain the anomaly detection result of the latent defect area.

[0138] The target visual attention area acquisition module 305 is used to acquire the user's target visual attention area and the equipment damage area of ​​the hidden defect area at the time of the accident trigger when the anomaly detection result is that damage has occurred.

[0139] The risk blind spot determination module 306 is used to determine the risk blind spot corresponding to the user based on the difference between the target visual attention area and the equipment damage area.

[0140] The technical solution of this invention acquires user eye gaze data in a virtual power grid scenario, determines the user's visual attention area based on the eye gaze data, generates a corresponding latent defect area in the user's visual attention area, and then performs damage detection on the latent defect area by combining equipment status parameters, material parameters, and user operation load data. This enables the defect setting in the virtual training process to be combined with the user's actual attention behavior. When damage occurs, by acquiring the user's target visual attention area and the equipment damage area, and determining the user's corresponding risk blind spot based on the difference between the two, it can identify areas that the user did not pay attention to but actually suffered damage, achieving accurate determination of the user's risk blind spot and improving the accuracy of risk identification analysis in the virtual power grid training process. This solves the technical problem of existing technologies that cannot identify risk blind spots formed by trainees not discovering potential risks, thus enabling the identification of risk blind spots formed by trainees not discovering potential risks.

[0141] In some embodiments, in determining the user's corresponding risk blind spot based on the difference between the target visual attention area and the device damage area, the risk blind spot determination module 306 is specifically used for:

[0142] The initial physical damage grid group is obtained by filtering based on the damage value corresponding to the equipment damage area and the preset failure threshold.

[0143] Connectivity analysis is performed on the spatial adjacency relationship between at least one grid cell in the initial physical damage grid group to obtain at least one damaged connected region;

[0144] Spatial overlap difference processing is performed on the damaged connected region and the target visual interest region to obtain the user's risk blind zone.

[0145] In some embodiments, in determining the user's initial visual attention region based on eye gaze data, the initial visual attention region determination module 302 is specifically used for:

[0146] Based on eye gaze data, determine at least one gaze point corresponding to the user;

[0147] Divide the virtual power grid scenario into at least one candidate region;

[0148] Based on the location coordinates of the fixation point, determine the number of fixations and fixation duration for the candidate region;

[0149] The fixation intensity of the candidate region is obtained by weighting the number of fixations and the fixation duration of the candidate region.

[0150] Based on at least one gaze intensity, determine the user's initial visual attention area from at least one candidate region.

[0151] In some embodiments, in determining a user's initial visual attention region from at least one candidate region based on at least one gaze intensity, the initial visual attention region determination module 302 is specifically configured to:

[0152] Calculate the fixation confidence factor for candidate regions;

[0153] Based on the gaze confidence factor of at least one candidate region and a preset attention threshold, at least one candidate region is filtered to determine the user's initial visual attention region.

[0154] In some embodiments, in obtaining anomaly detection results for the hidden defect area by performing damage detection on the power grid facilities in the hidden defect area based on equipment status parameters, material parameters, and operating load data, the anomaly detection result determination module 304 is specifically used for:

[0155] Based on the material parameters, determine the material mechanical parameters of the power grid facilities corresponding to the hidden defect areas;

[0156] Based on the equipment status parameters, determine the defect parameters corresponding to the latent defect areas;

[0157] Based on material mechanical parameters, defect parameters, and operating load data, physical evolution calculations are performed on the power grid facilities corresponding to the hidden defect areas to obtain stress distribution data, strain distribution data, and displacement data corresponding to the power grid facilities.

[0158] Based on stress distribution data, strain distribution data, and displacement data, the actual structural damage field corresponding to the hidden defect region is generated.

[0159] By comparing the damage degree parameters in the actual structural damage field with the preset accident triggering threshold, the abnormal detection results corresponding to the hidden defect area are obtained.

[0160] In some embodiments, after determining the risk blind spot corresponding to the user, the risk blind spot determination device for power grid training further includes:

[0161] Generate corresponding cognitive blind spot warning masks based on risk blind spots;

[0162] Obtain the real-time rendered screen that was locked at the moment the accident was triggered, and unfreeze the frozen graphics rendering thread;

[0163] Based on the cognitive blind spot warning mask and real-time rendering, the accident backtracks and re-enacts the evolution from the moment the accident was triggered to obtain the accident backtracking scenario corresponding to the risk blind spot;

[0164] Adjust the backtracking speed of the accident backtracking scenario based on the backtracking control commands input by the user;

[0165] The cognitive blind spot warning mask is overlaid on the rendering screen corresponding to the accident retrospective scene to show the hidden danger area corresponding to the risk blind spot and the damage evolution process of the hidden danger area.

[0166] In some embodiments, in generating a latent defect region corresponding to a virtual power grid scene within an initial visual attention region, the initial visual attention region determination module 302 is specifically used for:

[0167] Based on the gaze confidence factor, preset disclosure threshold and transition bandwidth parameter corresponding to the initial visual attention area, interval mapping calculation is performed to obtain the visual mixing coefficient matrix corresponding to at least one grid area in the 3D model of the virtual power grid scene.

[0168] Based on at least one visual blending coefficient matrix, an instantiated rendering indirect command buffer is generated. Based on the target mesh region in the instantiated rendering indirect command buffer where the blending coefficient is greater than a preset value, the corresponding micro-damage geometric displacement data and anisotropic crack normal data are matched from the latent defect model database to obtain a high-frequency texture resource package.

[0169] Based on the high-frequency texture resource package, the graphics processing unit performs texture decompression and video memory residency, and establishes a binding relationship between the texture resource identifier corresponding to the high-frequency texture resource package and the grid vertex coordinates in the 3D model of the virtual power grid scene;

[0170] Configure the defect visualization shader according to the binding relationship, and pass the local stress concentration coefficients corresponding to the hidden defect model into the defect visualization shader to obtain the active rendering context;

[0171] Pixel-level fusion rendering is performed based on the active rendering context, visual blending coefficient matrix, and preset rendering optimization instructions to generate hidden defect areas corresponding to the virtual power grid scene.

[0172] The power grid training risk blind spot determination device provided in this embodiment of the invention can execute the power grid training risk blind spot determination method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the power grid training risk blind spot determination method.

[0173] Figure 4 This is a structural schematic diagram of a risk blind spot determination device for power grid training provided in an embodiment of the present invention.

[0174] like Figure 4 As shown, the power grid training risk blind spot determination device 400 includes at least one processor 401 and a memory, such as a read-only memory (ROM) 402 or a random access memory (RAM) 403, communicatively connected to the at least one processor 401. The memory stores computer programs executable by the at least one processor. The processor 401 can perform various appropriate actions and processes based on the computer program stored in the ROM 402 or loaded from storage unit 408 into the RAM 403. The RAM 403 may also store various programs and data required for the operation of the power grid training risk blind spot determination device 400. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0175] The power grid training risk blind spot determination device 400 has multiple components connected to an I / O interface 405, including: an input unit 406, such as a keyboard, mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a hard disk, optical disk, etc.; and a communication unit 409, such as a network card, modem, wireless transceiver, etc. The communication unit 409 allows the power grid training risk blind spot determination device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0176] Processor 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 401 performs the various methods and processes described above, such as the risk blind spot determination method for power grid training.

[0177] In some embodiments, the power grid training risk blind spot determination method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the power grid training risk blind spot determination device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by processor 401, one or more steps of the power grid training risk blind spot determination method described above can be performed. Alternatively, in other embodiments, processor 401 can be configured to perform the power grid training risk blind spot determination method by any other suitable means (e.g., by means of firmware).

[0178] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0179] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0180] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0181] To provide user interaction, the systems and techniques described herein can be implemented on a power grid training blind spot determination device, which includes: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the power grid training blind spot determination device. Other types of devices can also be used to provide user interaction; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0182] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0183] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0184] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0185] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining risk blind spots in power grid training, characterized in that, The method includes: Acquire user eye gaze data in a virtual electric field scene; Based on the eye gaze data, the user's initial visual attention area is determined, and a hidden defect area corresponding to the virtual power grid scene is generated in the initial visual attention area. Obtain the equipment status parameters, material parameters, and user's operating load data within the area of ​​the hidden defect; Based on the equipment status parameters, the material parameters, and the operating load data, damage detection is performed on the power grid facilities in the latent defect area to obtain the abnormal detection results of the latent defect area. When the anomaly detection result indicates that damage has occurred, the user's target visual attention area at the time of the incident triggering, and the device damage area of ​​the hidden defect area are obtained. Based on the difference between the target visual attention area and the device damage area, the risk blind zone corresponding to the user is determined.

2. The method according to claim 1, characterized in that, The step of determining the risk blind spot corresponding to the user based on the difference between the target visual attention area and the device damage area includes: The initial physical damage grid group is obtained by filtering based on the damage value corresponding to the damaged area of ​​the equipment and the preset failure threshold. Connectivity analysis is performed on the spatial adjacency relationships between at least one grid cell in the initial physical damage grid group to obtain at least one damaged connected region; Spatial overlap difference processing is performed on the damaged connected region and the target visual attention region to obtain the user's risk blind zone.

3. The method according to claim 1, characterized in that, Determining the user's initial visual attention area based on the eye gaze data includes: Based on the eye gaze data, at least one gaze point corresponding to the user is determined; The virtual power grid scenario is divided into at least one candidate region; Based on the location coordinates of the fixation point, determine the number of fixations and the fixation duration of the candidate region; The gaze intensity of the candidate region is obtained by weighting the number of gazes and the gaze duration of the candidate region. The user's initial visual attention region is determined from at least one of the candidate regions based on at least one of the gaze intensities.

4. The method according to claim 3, characterized in that, Determining the user's initial visual attention region from at least one of the candidate regions based on at least one of the gaze intensities includes: Calculate the gaze confidence factor of the candidate region; Based on the gaze confidence factor of at least one of the candidate regions and a preset attention threshold, at least one of the candidate regions is filtered to determine the user's initial visual attention region.

5. The method according to claim 1, characterized in that, The step of performing damage detection on the power grid facilities in the latent defect area based on the equipment status parameters, the material parameters, and the operating load data to obtain the abnormal detection results of the latent defect area includes: Based on the material parameters, determine the material mechanical parameters of the power grid facilities corresponding to the hidden defect area; Based on the equipment status parameters, determine the defect parameters corresponding to the latent defect area; Based on the material mechanical parameters, the defect parameters, and the operating load data, physical evolution calculations are performed on the power grid facilities corresponding to the hidden defect region to obtain stress distribution data, strain distribution data, and displacement data corresponding to the power grid facilities. Based on the stress distribution data, the strain distribution data, and the displacement data, the actual structural damage field corresponding to the hidden defect region is generated; The abnormal detection result corresponding to the hidden defect area is obtained by comparing the damage degree parameter in the real structural damage field with the preset accident triggering threshold.

6. The method according to claim 1, characterized in that, After determining the risk blind spot corresponding to the user, the method further includes: Generate a corresponding cognitive blind spot warning mask based on the aforementioned risk blind spots; Obtain the real-time rendered screen that was locked at the moment the accident was triggered, and unfreeze the frozen graphics rendering thread; Based on the cognitive blind spot warning mask and the real-time rendered image, the accident backtracking and evolution is performed backward from the moment the accident was triggered to obtain the accident backtracking scene corresponding to the risk blind spot; Adjust the backtracking speed of the accident backtracking scenario according to the backtracking control command input by the user; The cognitive blind spot warning mask is overlaid on the rendering screen corresponding to the accident retrospective scene to display the hidden danger area corresponding to the risk blind spot and the damage evolution process of the hidden danger area.

7. The method according to claim 1, characterized in that, The step of generating the latent defect region corresponding to the virtual power grid scene in the initial visual attention region includes: Based on the gaze confidence factor, preset disclosure threshold and transition bandwidth parameter corresponding to the initial visual attention region, interval mapping calculation is performed to obtain the visual mixing coefficient matrix corresponding to at least one grid region in the three-dimensional model of the virtual power grid scene. Based on at least one of the visual blending coefficient matrices, an instantiated rendering indirect command buffer is generated. Based on the target mesh region in the instantiated rendering indirect command buffer where the blending coefficient is greater than a preset value, the corresponding micro-damage geometric displacement data and anisotropic crack normal data are matched from the latent defect model database to obtain a high-frequency texture resource package. Based on the high-frequency texture resource package, the graphics processing unit performs texture decompression and video memory resident, and establishes a binding relationship between the texture resource identifier corresponding to the high-frequency texture resource package and the grid vertex coordinates in the three-dimensional model of the virtual power grid scene; Configure the defect visualization shader according to the binding relationship, and pass the local stress concentration coefficient corresponding to the hidden defect model into the defect visualization shader to obtain the active rendering context; Pixel-level fusion rendering is performed based on the active rendering context, the visual blending coefficient matrix, and preset rendering optimization instructions to generate the hidden defect area corresponding to the virtual power grid scene.

8. A device for determining risk blind spots in power grid training, characterized in that, include: The eye gaze data acquisition module is used to acquire the user's eye gaze data in the virtual electric field scene; The initial visual attention area determination module is used to determine the user's initial visual attention area based on the eye gaze data, and generate a hidden defect area corresponding to the virtual power grid scene in the initial visual attention area. The latent defect area data acquisition module is used to acquire equipment status parameters, material parameters, and user operating load data within the latent defect area; An anomaly detection result determination module is used to perform damage detection on the power grid facilities in the latent defect area based on the equipment status parameters, the material parameters and the operating load data, and obtain the anomaly detection result of the latent defect area; The target visual attention area acquisition module is used to acquire the user's target visual attention area and the equipment damage area of ​​the hidden defect area at the time of the accident triggering when the anomaly detection result is that damage has occurred. The risk blind spot determination module is used to determine the risk blind spot corresponding to the user based on the difference between the target visual attention area and the device damage area.

9. A device for determining risk blind spots in power grid training, characterized in that, The equipment for identifying risk blind spots in power grid training includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the risk blind spot determination method for power grid training as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for determining risk blind spots in power grid training as described in any one of claims 1-7.

Images