Information processing method, device and system of power grid digital system and storage medium
By dividing the target area in the power grid digital system and selecting a matching identification model, automated fault detection is performed on completed infrastructure resources, solving the problem of large manual field workload in existing technologies and improving the efficiency of automated management and monitoring of power grid infrastructure resources.
Patent Information
- Application Number
- CN202510553140.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-29
- Publication Date
- 2025-09-19
AI Technical Summary
The existing power grid infrastructure information processing method is mainly concentrated in the construction phase, lacking automated management and monitoring of completed infrastructure resources, resulting in a large workload of manual field work and low efficiency.
The infrastructure area to be inspected is divided into target areas based on geographical environment information and meteorological information. A matching recognition model is selected to automatically identify infrastructure resource faults in the target image. The image is collected and preprocessed by image shooting equipment, and accurate detection is performed using the target recognition model.
It reduces the interference of geographical environment and meteorological information differences on the identification model, improves the accuracy of fault detection and the efficiency of automated monitoring and management of completed infrastructure resources, and reduces the workload of manual field work.
Smart Images
Figure CN120672012A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of power grid infrastructure information processing, and in particular to an information processing method, device, system and storage medium for a power grid digitalization system. Background Art
[0002] Power grid infrastructure construction spans a large region. As the scale of the power grid continues to expand, the amount of grid infrastructure required is increasing. Therefore, to achieve centralized operation and dispatch of power grid infrastructure projects, grid infrastructure information needs to be integrated into a digital grid system, allowing it to be processed. Current methods for processing grid infrastructure information primarily focus on the construction phase, lacking the ability to process information related to completed infrastructure resources. This requires significant manual fieldwork, hindering the automated management and monitoring of completed infrastructure resources by the digital grid system. Summary of the Invention
[0003] The present invention provides an information processing method, device, and storage medium for a power grid digital system to at least address the low settlement efficiency issue in related technologies. The technical solution of the present invention is as follows:
[0004] According to a first aspect of an embodiment of the present invention, there is provided an information processing method for a power grid digitization system, which is applied to the power grid digitization system. The method comprises: dividing an infrastructure area to be inspected into multiple target areas according to geographic environment information and meteorological information, and determining target geographic environment information and target meteorological information for each target area; determining a target recognition model that matches the target geographic environment information and target meteorological information from multiple image recognition models; the multiple image recognition models correspond to different geographic environment information and meteorological information, respectively; inputting target images of each target area into corresponding target recognition models, identifying faults of target infrastructure resources in each corresponding target image, and obtaining target recognition results for each target area.
[0005] In one implementation, a power grid digitalization system includes an image capture device, which includes a flying device, an airborne digital camera, and a lidar scanner. The method further includes: for any target area, controlling the image capture device to capture multiple acquired images of target geographic environment information and target meteorological information of any target area according to a preset route and preset shooting angle corresponding to any target area; performing a preprocessing operation on the multiple acquired images, and determining a target image that meets image index conditions from the multiple acquired images after the preprocessing operation; wherein the preprocessing operation includes one or more of the following: image cropping, image scaling, resolution adjustment, and grayscale adjustment.
[0006] In one implementation, a target image that meets an image index condition is determined from a plurality of acquired images after a preprocessing operation, including: determining an image difference between the plurality of acquired images after the preprocessing operation and a standard image; determining an image that meets the image index condition from the plurality of acquired images after the preprocessing operation as a candidate image; and selecting an image whose image difference is greater than a preset difference from the candidate images as the target image; the standard image is an image obtained by shooting according to a preset route and preset shooting angle corresponding to any target area and performing a preprocessing operation.
[0007] In one implementation, the target images of each target area are respectively input into the corresponding target recognition model, and the faults of the target infrastructure resources in the corresponding target images are identified to obtain the target recognition results of each target area; including: for any target image of any target area, using the target recognition model matched with any target area, extracting features of the target object representing the target infrastructure resource in any target image, and analyzing and identifying whether the features of the target object meet the fault features; analyzing and identifying to determine whether the target fault exists in the target infrastructure resource, and locating the fault position and fault type of the target fault in the target image; or analyzing and identifying to determine whether the target infrastructure resource does not have a fault.
[0008] In one implementation, the target recognition model includes a target model classifier and multiple target sub-models, the target model classifier is used to identify the target infrastructure resources of the target image, and the target sub-model is used to identify the fault of a target infrastructure resource; the target images of each target area are respectively input into the corresponding target recognition model, and the faults of the target infrastructure resources in the corresponding target images are identified to obtain target recognition results for each target area, including: for any target image of any target area, calling the target model classifier to determine at least one target infrastructure resource included in any target image; the target sub-model corresponds to the infrastructure resource one-to-one; according to the one-to-one correspondence between the target sub-model and the infrastructure resource, determining each target sub-model corresponding to each target infrastructure resource in at least one target infrastructure resource; inputting any target image into each target sub-model, calling each target sub-model to identify the fault type and fault location of the corresponding target infrastructure resource, so as to determine the recognition result output by each target sub-model as the target recognition result.
[0009] In one implementation, the method further includes: training each preset recognition model under different geographical environment information and meteorological information respectively to obtain multiple image recognition models; wherein, the training process for any preset recognition model includes: collecting multiple groups of images of the same infrastructure area with the same geographical environment information and the same meteorological information; the multiple groups of images include fault marking results, and the fault marking results include fault type markings and fault location markings; inputting the multiple groups of images into the preset recognition model to obtain the model prediction output results of the preset recognition model, and based on the loss value between the model prediction output results and the fault marking results, adjusting the model parameters of the preset model to obtain the image recognition model under the condition of convergence of the loss value.
[0010] In one implementation, the target infrastructure resources include one or more of the following: poles, ground wires, and insulator strings. The method further includes: determining a target time period to which the system access time of a user account accessing the power grid digital system belongs and a target region to which the user account belongs; determining a target control level corresponding to the target time period and target region based on a correspondence between the system access time period and the account region and the access control security level; determining a target login method for the user account to log into the power grid digital system based on a target access operation corresponding to the target control level; the target control level is positively correlated with the operational complexity of the target login method.
[0011] According to a second aspect of an embodiment of the present invention, an information processing device for a power grid digitization system is provided, and the information processing device for the power grid digitization system includes: a division unit, used to divide an infrastructure area to be detected into multiple target areas according to geographical environment information and meteorological information, and determine the target geographical environment information and target meteorological information of each target area; a determination unit, used to determine a target recognition model that matches the target geographical environment information and target meteorological information from multiple image recognition models; the multiple image recognition models correspond to different geographical environment information and meteorological information, respectively; and an identification unit, used to input the target images of each target area into the corresponding target recognition model, identify the faults of the target infrastructure resources in the corresponding target images, and obtain target recognition results for each target area.
[0012] According to a third aspect of an embodiment of the present invention, there is provided an information processing device for a power grid digitalization system, which has a shooting function and is configured to execute the information processing method for a power grid digitalization system as described in the first aspect and any possible implementation thereof.
[0013] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which instructions are stored. When the instructions in the computer-readable storage medium are executed by a processor of a power control device, the power control device is enabled to execute the information processing method of a power grid digitization system as described in the first aspect and any possible implementation thereof.
[0014] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, which includes computer instructions. When the computer instructions are executed on a power control device, the power control device executes the information processing method of the power grid digitization system of the above-mentioned first aspect and any possible implementation thereof.
[0015] The technical solution provided by the embodiments of the present invention brings at least the following beneficial effects: in the infrastructure area to be detected, the infrastructure areas with similar geographical environment information and meteorological information are divided into the same target area, and a target recognition model that matches the geographical environment information and meteorological information of the target area is selected, and whether the target infrastructure resources in the target image of the target area have faults is automatically and accurately identified, so as to reduce the interference of the difference in geographical environment information and meteorological information on the image recognition model, improve the power grid digital system's accurate detection of the fault status of completed infrastructure resources, reduce the workload of manual field work, and improve the power grid digital system's automated monitoring and management efficiency of completed infrastructure resources.
[0016] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.
[0018] Figure 1 is a flow chart showing an information processing method for a power grid digital system according to an exemplary embodiment;
[0019] Figure 2 is a block diagram showing an information processing device for a power grid digitalization system according to an exemplary embodiment;
[0020] Figure 3 is a schematic diagram showing a power control device according to an exemplary embodiment. DETAILED DESCRIPTION
[0021] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
[0022] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.
[0023] Before introducing in detail the information processing method of the power grid digital system provided by the embodiment of the present application, a brief introduction to the application scenarios and implementation environment involved in the embodiment of the present application is first given.
[0024] First, a brief introduction to the application scenarios involved in this application is given.
[0025] Power grid infrastructure construction spans a large region. As the scale of the power grid continues to expand, the amount of grid infrastructure required is increasing. Therefore, to achieve centralized operation and dispatch of power grid infrastructure projects, grid infrastructure information needs to be integrated into a digital grid system, allowing it to be processed. Current methods for processing grid infrastructure information primarily focus on the construction phase, lacking the ability to process information related to completed infrastructure resources. This requires significant manual fieldwork, hindering the automated management and monitoring of completed infrastructure resources by the digital grid system.
[0026] In response to the above problems, the present application proposes an information processing method for a power grid digitalization system, which divides the infrastructure areas to be detected that have similar geographical environment information and meteorological information into the same target area, and selects a target recognition model that matches the geographical environment information and meteorological information of the target area, and automatically and accurately identifies whether a target infrastructure resource in the target image of the target area has a fault, so as to reduce the interference of the difference between the geographical environment information and the meteorological information on the image recognition model, improve the power grid digitalization system's accurate detection of the fault status of completed infrastructure resources, reduce the workload of manual field work, and improve the power grid digitalization system's efficiency in automated monitoring and management of completed infrastructure resources.
[0027] For ease of understanding, the information processing method of the power grid digitalization system provided by this application is specifically introduced below with reference to the accompanying drawings.
[0028] Figure 1 FIG. 1 is a flow chart showing an information processing method for a power grid digital system according to an exemplary embodiment. Figure 1As shown, the information processing method of the power grid digital system includes the following steps.
[0029] S11 , dividing the infrastructure area to be inspected into multiple target areas according to the geographical environment information and the meteorological information, and determining the target geographical environment information and the target meteorological information of each target area.
[0030] S12, determining a target recognition model that matches the target geographic environment information and the target meteorological information from a plurality of image recognition models.
[0031] Multiple image recognition models correspond to different geographical environment information and meteorological information.
[0032] S13 , inputting the target images of the respective target areas into corresponding target recognition models, identifying the faults of the target infrastructure resources in the corresponding target images, and obtaining target recognition results for the respective target areas.
[0033] The above-mentioned target infrastructure resources include one or more of the following: poles, towers, ground wires and insulator strings.
[0034] Optionally, a deep learning algorithm is used to train a large number of normal and abnormal images of poles, ground wires, and insulator strings, so that the model can automatically identify quality problems in the images, such as cracks in poles, broken strands in ground wires, and damaged insulator strings.
[0035] Through the above implementation method, the infrastructure areas to be detected that have similar geographical environment information and meteorological information are divided into the same target area, and a target recognition model that matches the geographical environment information and meteorological information of the target area is selected to automatically and accurately identify whether the target infrastructure resources in the target image of the target area have failed, so as to reduce the interference of the difference in geographical environment information and meteorological information on the image recognition model, improve the power grid digital system's accurate detection of the fault status of completed infrastructure resources, reduce the workload of manual field work, and improve the power grid digital system's automated monitoring and management efficiency of completed infrastructure resources.
[0036] As an embodiment, the power grid digitization system includes an image capture device, which includes a flying device, an airborne digital camera, and a lidar scanner.
[0037] Based on this, in order to improve the adaptability of the target image and the model: for any target area, control the image capture equipment, and collect multiple collected images of the target geographical environment information and target meteorological information of any target area according to the preset route and preset shooting angle corresponding to any target area; perform preprocessing operations on the multiple collected images, and determine the target image that meets the image index conditions from the multiple collected images after the preprocessing operation; wherein the preprocessing operations include one or more of the following: image cropping, image scaling, resolution adjustment, and grayscale adjustment.
[0038] The above-mentioned flying device can be a drone.
[0039] Correspondingly, in another embodiment, an image that meets the image index condition is selected from a plurality of collected images, and then a preprocessing operation is performed on the image that meets the image index condition to obtain a target image.
[0040] In this embodiment, the image quality of the input target recognition model is improved by screening the image indicators of the captured image; and the image is preprocessed to make the input target image compatible with the image requirements recognized by the target recognition model, so that the target recognition model can more accurately recognize the target image and improve the accuracy of the recognition result.
[0041] Optionally, in order to reduce invalid calls of the target recognition model, non-fault images taken along a preset route and a preset shooting angle when no fault occurs are preprocessed to obtain standard images, and the standard images are stored in the power grid digitization system. Among the collected images after the preprocessing operation, images that differ greatly from the standard images are determined as target images to avoid images that differ little from the standard images from being input into the image recognition model, thereby performing invalid recognition of non-fault conditions.
[0042] A large difference indicates a high probability of a fault, while a small difference indicates a low probability of a fault. If the difference between the acquired image after preprocessing and the standard image is small, it is determined that the target image does not exist, and there is no need to call the target recognition model, thus reducing the number of invalid calls.
[0043] Correspondingly, a target image that meets the image index conditions is determined from the multiple acquired images after the preprocessing operation. The specific implementation steps include the following: Determining the image difference between the multiple acquired images after the preprocessing operation and the standard image. Images that meet the image index conditions among the multiple acquired images after the preprocessing operation are determined as candidate images; Images whose image difference is greater than a preset difference are selected from the candidate images as target images; the standard image is an image obtained by capturing a target area along a preset route and at a preset shooting angle and performing the preprocessing operation.
[0044] The target images of each target area are respectively input into the corresponding target recognition model, and the faults of the target infrastructure resources in the corresponding target images are identified to obtain the target recognition results of each target area; the specific process is as follows.
[0045] First, for any target image of any target area, a target recognition model matching any target area is used to extract features of the target object representing the target infrastructure resources in any target image, and analyze and identify whether the features of the target object meet the fault features.
[0046] Secondly, analyzing and identifying whether a target fault exists in the target infrastructure resource, and locating the fault position and fault type of the target fault in the target image; or analyzing and identifying whether a target infrastructure resource does not exist in the target infrastructure resource.
[0047] In the above embodiment, the target recognition model is used to directly identify and locate image faults. That is, the corresponding model training process is as follows: images with infrastructure resource faults in the sample image group are annotated to indicate the type and location of the faults in the images.
[0048] As an implementation method, the target recognition model includes a target model classifier and multiple target sub-models. The target model classifier is used to identify target infrastructure resources in a target image, and the target sub-model is used to identify a fault of a target infrastructure resource.
[0049] Based on this implementation, the target images of each target area are respectively input into the corresponding target recognition model, the faults of the target infrastructure resources in the corresponding target images are identified, and the target recognition results of each target area are obtained. The specific implementation process is as follows.
[0050] First, for any target image in any target area, a target model classifier is called to determine at least one target infrastructure resource included in any target image.
[0051] The target sub-model corresponds one-to-one to the infrastructure resources.
[0052] Secondly, according to the one-to-one correspondence between the target sub-models and the infrastructure resources, each target sub-model corresponding to each target infrastructure resource in the at least one target infrastructure resource is determined.
[0053] Third, any target image is input into each target sub-model, and each target sub-model is called to identify the fault type and fault location of a corresponding target infrastructure resource, so that the recognition result output by each target sub-model is determined as the target recognition result.
[0054] In this implementation, in order to avoid mutual interference between infrastructure resources during the fault identification process, a sub-model is set up for each infrastructure resource, so that each sub-model can more accurately identify the fault type and fault location of only one infrastructure resource, thereby improving the accuracy of the target identification results.
[0055] As an implementation method, each preset recognition model under different geographical environment information and meteorological information is trained respectively to obtain multiple image recognition models.
[0056] The training process for any preset recognition model includes the following steps.
[0057] First, multiple sets of images of the same infrastructure area with the same geographical environment information and the same meteorological information are collected. The multiple sets of images include fault marking results, which include fault type markings and fault location markings.
[0058] Secondly, multiple groups of images are input into the preset recognition model to obtain the model prediction output results of the preset recognition model. Based on the loss value between the model prediction output results and the fault marking results, the model parameters of the preset model are adjusted to obtain the image recognition model under the condition of loss value convergence.
[0059] In the above-described embodiment, the training process of the sub-recognition model is generally similar to the training process of the image recognition model. The differences between the two are as follows: multiple groups of images are first classified according to infrastructure resource type, and the multiple groups of images only include the fault type identification and fault location identification of one infrastructure resource, thereby obtaining multiple groups of sub-model image samples corresponding to multiple different infrastructure resource types. For any group of sub-model image samples, the sample includes the identification of the infrastructure resource type. A sub-recognition model that determines the fault type and fault location of the corresponding infrastructure resource type using any group of sub-model image samples is trained to obtain a trained sub-recognition model for application.
[0060] As an implementation method, a target time period to which the system access time of a user account accessing the power grid digital system belongs and a target region to which the user account belongs are determined; based on the correspondence between the system access time period and the account region and the access control security level, a target control level corresponding to the target time period and the target region is determined; based on the target access operation corresponding to the target control level, a target login method for the user account to log into the power grid digital system is determined; the size of the target control level is positively correlated with the operational complexity of the target login method.
[0061] In this implementation, different encryption methods are set for different system access time periods. For example, during daytime hours, more people participate in the workflow and their reliability is higher than during nighttime hours. Therefore, the security level for daytime hours is set lower than that for nighttime hours. Regions with high regional security are set to a lower access control security level than regions with low regional security. Regional security is negatively correlated with the access control security level.
[0062] Through this implementation, the complexity of access methods to the power grid digital system is divided by time and region, so as to improve the security of access to the power grid digital system, thereby avoiding malicious operations on data in the power grid digital system (such as multiple stored image recognition models) and improving the reliability of the power grid digital system.
[0063] In order to realize the above functions, the information processing device of the power grid digitalization system includes hardware structures and / or software modules corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0064] The present disclosure also provides a Figure 2 The information processing device of the power grid digital system shown includes: a dividing unit 21, a determining unit 22 and an identifying unit 23.
[0065] The division unit 21 is used to divide the infrastructure area to be inspected into multiple target areas according to the geographical environment information and the meteorological information, and determine the target geographical environment information and the target meteorological information of each target area.
[0066] The determination unit 22 is used to determine a target recognition model that matches the target geographical environment information and the target meteorological information from a plurality of image recognition models; the plurality of image recognition models correspond to different geographical environment information and meteorological information respectively.
[0067] The recognition unit 23 is configured to input the target images of each target area into the corresponding target recognition model, recognize the faults of the target infrastructure resources in each corresponding target image, and obtain target recognition results for each target area.
[0068] In one embodiment, a power grid digitization system includes an image capture device, which includes a flying device, an airborne digital camera, and a lidar scanner; the device is further used to: for any target area, control the image capture device to capture multiple acquired images of target geographic environment information and target meteorological information of any target area according to a preset route and preset shooting angle corresponding to any target area; perform a preprocessing operation on the multiple acquired images, and determine a target image that meets image index conditions from the multiple acquired images after the preprocessing operation; wherein the preprocessing operation includes one or more of the following: image cropping, image scaling, resolution adjustment, and grayscale adjustment.
[0069] In one embodiment, the determination unit 22 is used to: determine the image difference between multiple acquired images after the preprocessing operation and the standard image; determine the images that meet the image index conditions among the multiple acquired images after the preprocessing operation as candidate images; select images whose image difference is greater than a preset difference from the candidate images as target images; the standard image is an image obtained by shooting according to a preset route and preset shooting angle corresponding to any target area and performing a preprocessing operation.
[0070] In one embodiment, the recognition unit 23 is used to: for any target image of any target area, use a target recognition model matched with any target area to extract features of a target object representing a target infrastructure resource in any target image, and analyze and identify whether the features of the target object meet the fault features; analyze and identify to determine whether a target fault exists in the target infrastructure resource, and locate the fault position and fault type of the target fault in the target image; or analyze and identify to determine whether there is no fault in the target infrastructure resource.
[0071] In one embodiment, the target recognition model includes a target model classifier and multiple target sub-models. The target model classifier is used to identify target infrastructure resources in a target image, and the target sub-model is used to identify faults in a target infrastructure resource. The recognition unit 23 is specifically configured to: for any target image in any target area, call the target model classifier to determine at least one target infrastructure resource included in any target image; the target sub-models correspond one-to-one with the infrastructure resources; based on the one-to-one correspondence between the target sub-models and the infrastructure resources, determine each target sub-model corresponding to each target infrastructure resource in the at least one target infrastructure resource; input any target image into each target sub-model, call each target sub-model to identify the fault type and fault location of the corresponding target infrastructure resource, and determine the recognition results output by each target sub-model as the target recognition result.
[0072] In one embodiment, the device is also used to: train each preset recognition model under different geographical environment information and meteorological information respectively to obtain multiple image recognition models; wherein, the training process for any preset recognition model includes: collecting multiple groups of images of the same infrastructure area with the same geographical environment information and the same meteorological information; the multiple groups of images include fault marking results, and the fault marking results include fault type markings and fault location markings; inputting the multiple groups of images into the preset recognition model to obtain the model prediction output results of the preset recognition model, and based on the loss value between the model prediction output results and the fault marking results, adjusting the model parameters of the preset model to obtain the image recognition model under the condition of convergence of the loss value.
[0073] In one embodiment, the target infrastructure resources include one or more of the following: poles, ground wires, and insulator strings. The device is also used to: determine the target time period to which the system access time of a user account accessing the power grid digital system belongs and the target region to which the user account belongs; determine the target control level corresponding to the target time period and the target region based on the correspondence between the system access time period and the account region and the access control security level; determine the target login method for the user account to log into the power grid digital system based on the target access operation corresponding to the target control level; the size of the target control level is positively correlated with the operational complexity of the target login method.
[0074] Regarding the device in the above embodiment, the specific manner in which each unit module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.
[0075] Figure 3 This is a schematic diagram of a power control device provided by this application. Figure 3 The power control device 50 may include at least one processor 501 and a memory 503 for storing processor-executable instructions. The processor 501 is configured to execute instructions in the memory 503 to implement the information processing method of the power grid digitalization system in the following embodiments.
[0076] In addition, the power control device 50 may further include a communication bus 502 , at least one communication interface 504 , an input device 506 , and an output device 505 .
[0077] The processor 501 may be a central processing unit (CPU), a microprocessor unit, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.
[0078] The communication bus 502 may include a pathway for transmitting information between the aforementioned components.
[0079] The communication interface 504 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.
[0080] The input device 506 is used to receive input signals and the output device 505 is used to output signals.
[0081] The memory 503 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be independent and connected to the processing unit via a bus. The memory may also be integrated with the processing unit.
[0082] The memory 503 is used to store instructions for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the instructions stored in the memory 503, thereby realizing the functions of the method of the present application.
[0083] In a specific implementation, as an embodiment, the processor 501 may include one or more CPUs, such as Figure 3 CPU0 and CPU1 in.
[0084] In a specific implementation, as an embodiment, the power control device 50 may include multiple processors, such as Figure 3 1 and 507. Each of these processors may be a single-CPU processor or a multi-CPU processor. A processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).
[0085] The power control device is Figure 3 The system includes a processor 501 and a memory 503 for storing executable instructions for the processor 501. The processor 501 is configured to execute the executable instructions to implement the information processing method for a power grid digital system according to any of the above-described possible implementations. The methods can achieve the same technical effects and are not described here in detail to avoid repetition.
[0086] The present application also provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an information processing device or power control device in a digital power grid system, the information processing device or power control device in the digital power grid system can execute the information processing method for a digital power grid system according to any of the possible implementations described above. The same technical effects can be achieved, and to avoid repetition, they are not described here in detail.
[0087] The present application also provides a computer program product including a computer program or instructions. The computer program or instructions are executed by a processor to implement the information processing method for a power grid digitalization system according to any of the above-described possible implementations. The method achieves the same technical effects and is not further described here to avoid repetition.
[0088] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.
[0089] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. An information processing method for a power grid digital system, characterized in that: Applied to a power grid digital system, the method includes: Divide the infrastructure area to be inspected into multiple target areas according to the geographical environment information and meteorological information, and determine the target geographical environment information and target meteorological information of each target area; Determining a target recognition model that matches the target geographical environment information and the target meteorological information from a plurality of image recognition models, wherein the plurality of image recognition models correspond to different geographical environment information and meteorological information respectively; The target images of each target area are respectively input into the corresponding target recognition model, and the faults of the target infrastructure resources in each corresponding target image are identified to obtain the target recognition results of each target area.
2. The information processing method of the power grid digital system according to claim 1, characterized in that: The power grid digitization system includes an image capture device, which includes a flying device, an airborne digital camera, and a laser radar scanner; the method further includes: For any target area, the image capturing device is controlled to capture multiple acquired images of the target geographical environment information and the target meteorological information of the target area according to the preset route and preset shooting angle corresponding to the target area; a preprocessing operation is performed on the multiple acquired images, and the target image that meets the image index conditions is determined from the multiple acquired images after the preprocessing operation; wherein the preprocessing operation includes one or more of the following: image cropping, image scaling, resolution adjustment and grayscale adjustment.
3. The information processing method of the power grid digital system according to claim 2, characterized in that: Determining the target image that meets the image index condition from the plurality of collected images after the preprocessing operation includes: determining image differences between the plurality of acquired images after the preprocessing operation and the standard image; Among the multiple acquired images after the preprocessing operation, images that meet the image index conditions are determined as candidate images; images whose image differences are greater than preset differences are selected from the candidate images as the target images; the standard image is an image obtained by shooting according to the preset route and the preset shooting angle corresponding to any target area and performing the preprocessing operation.
4. The information processing method of the power grid digital system according to claim 1, characterized in that: Inputting the target images of the respective target areas into the corresponding target recognition models, identifying the faults of the target infrastructure resources in the corresponding target images, and obtaining target recognition results for the respective target areas; comprising: For any target image of any target area, using the target recognition model matched with the any target area, extracting features of the target object representing the target infrastructure resource in the any target image, and analyzing and identifying whether the features of the target object meet the fault features; Analyzing and identifying to determine that a target fault exists in the target infrastructure resource, and locating the fault position and fault type of the target fault in the target image; or analyzing and identifying to determine that no fault exists in the target infrastructure resource.
5. The information processing method of the power grid digital system according to claim 1, characterized in that: The target recognition model includes a target model classifier and multiple target sub-models. The target model classifier is used to identify the target infrastructure resources in the target image, and the target sub-model is used to identify the fault of a target infrastructure resource. The target images of each target area are respectively input into the corresponding target recognition model, and the fault of the target infrastructure resource in each corresponding target image is identified to obtain the target recognition result of each target area, including: For any target image of any target area, calling the target model classifier to determine at least one target infrastructure resource included in the target image; the target sub-model corresponds to the infrastructure resource in a one-to-one manner; Determining, according to a one-to-one correspondence between the target sub-models and the infrastructure resources, respective target sub-models corresponding to respective target infrastructure resources in the at least one target infrastructure resource; Input any one of the target images into each of the target sub-models, call each of the target sub-models to identify the fault type and fault location of a corresponding target infrastructure resource, and determine the recognition result output by each of the target sub-models as the target recognition result.
6. The information processing method of the power grid digital system according to any one of claims 1 to 5, characterized in that: The method further comprises: Training each preset recognition model under different geographical environment information and meteorological information respectively to obtain the multiple image recognition models; Among them, the training process for any preset recognition model includes: collecting multiple groups of images of the same infrastructure area with the same geographical environment information and the same meteorological information; the multiple groups of images include fault marking results, and the fault marking results include fault type markings and fault location markings; inputting the multiple groups of images into the preset recognition model to obtain the model prediction output results of the preset recognition model, and based on the loss value between the model prediction output results and the fault marking results, adjusting the model parameters of the preset model to obtain the image recognition model under the condition that the loss value converges.
7. The information processing method of a power grid digital system according to any one of claims 1 to 5, characterized in that: The target infrastructure resources include one or more of the following: poles, ground wires, and insulator strings. The method further includes: Determine a target time period to which a system access time of a user account accessing the power grid digital system belongs and a target region to which the user account belongs; Determine the target control level corresponding to the target time period and the target region based on the correspondence between the system access time period, account region, etc. and the access control security level; According to the target access operation corresponding to the target control level, the target login method for the user account to log into the power grid digitalization system is determined; the size of the target control level is positively correlated with the operation complexity of the target login method.
8. An information processing device for a power grid digital system, characterized in that: Applied to a power grid digital system, the device comprises: a division unit, configured to divide the infrastructure area to be inspected into a plurality of target areas according to the geographical environment information and the meteorological information, and to determine the target geographical environment information and the target meteorological information of each of the target areas; a determination unit, configured to determine a target recognition model that matches the target geographical environment information and the target meteorological information from a plurality of image recognition models; the plurality of image recognition models respectively corresponding to different geographical environment information and meteorological information; The recognition unit is used to input the target images of each target area into the corresponding target recognition model, identify the faults of the target infrastructure resources in the corresponding target images, and obtain the target recognition results of each target area.
9. A power grid digitalization system, characterized in that: The invention comprises an image capturing device, wherein the image capturing device comprises a flying device, an airborne digital camera and a laser radar scanner; and is configured to execute the information processing method of the power grid digitization system according to any one of claims 1 to 7.
10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of a power control device, the power control device is enabled to execute the information processing method for a power grid digitalization system according to any one of claims 1 to 7.