A method, apparatus, electronic device, and storage medium for identifying the status of a disconnect switch.

Abstract

This application relates to a method, apparatus, electronic device, and storage medium for identifying the status of a disconnect switch. The method includes: detecting an image to be identified according to a preset first target detection model to obtain the disconnect switch target boundary, disconnect switch type, and basic open / closed state of the disconnect switch in the image to be identified; obtaining a disconnect switch target image containing only the disconnect switch from the image to be identified based on the disconnect switch target boundary; detecting the positional relationship between the disconnect switch arm in the disconnect switch target image and preset disconnect switch structural components corresponding to the disconnect switch arm; and determining the status of the disconnect switch based on the basic open / closed state, disconnect switch type, and positional relationship. This application improves the accuracy of disconnect switch status identification by utilizing the disconnect switch type, initial closed state, and positional relationship to determine the final state of the disconnect switch.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, electronic device, and storage medium for identifying the status of a switch. Background Technology

[0002] In the inspection of smart substations, the status identification of disconnectors is a crucial step. Traditional manual inspections are time-consuming, labor-intensive, and inefficient. Existing image processing methods for disconnector status identification mostly rely on traditional image processing techniques or require manually setting calibration lines and obtaining parameters from on-site cameras. Traditional image processing techniques for disconnector arm identification lack robustness in complex backgrounds and with significant lighting variations. Manually setting calibration lines and obtaining on-site camera parameters reduces the level of automation, is cumbersome and complex, and requires specialized experience from personnel. Therefore, existing disconnector status identification technologies suffer from low accuracy.

[0003] There is currently no effective solution to the problem of low accuracy in the identification results of related technologies. Summary of the Invention

[0004] This embodiment provides a method, apparatus, electronic device, and storage medium for identifying the status of a disconnect switch, in order to solve the problem of low accuracy of identification results in related technologies.

[0005] Firstly, this embodiment provides a method for identifying the status of a disconnect switch, the method comprising:

[0006] According to the preset first target detection model, the image to be identified containing the disconnect switch is detected to obtain the disconnect switch type and basic open / closed state of the disconnect switch in the image to be identified;

[0007] Based on the image to be identified, the positional relationship between the knife arm and the knife switch structural component is determined; the knife switch structural component is pre-configured for the knife arm;

[0008] The state of the disconnector is determined based on the basic open / closed state of the disconnector, the disconnector type, and the positional relationship.

[0009] In some embodiments, determining the positional relationship between the knife arm and the knife switch structural component based on the image to be identified includes:

[0010] From the image to be identified, a target image of the disconnect switch containing the disconnect switch is obtained; the area of ​​the disconnect switch in the target image of the disconnect switch is greater than the area ratio threshold.

[0011] The step of determining the positional relationship between the knife arm and the knife switch structural components based on the image to be identified includes:

[0012] The knife arm and the knife gate structure in the target image of the knife gate are detected, and the positional relationship between the knife arm and the preset knife gate structure is determined based on the detection results.

[0013] In some embodiments, when detecting an image to be identified containing a disconnector to obtain the disconnector type and basic open / closed state of the disconnector in the image to be identified, the method further includes:

[0014] According to the preset first target detection model, the image to be identified, which contains the knife switch, is detected to obtain the knife switch target boundary in the image to be identified;

[0015] Obtaining the target image of the disconnector containing the disconnector from the image to be identified includes:

[0016] Based on the target boundary of the disconnector, an image of the disconnector target containing only the disconnector is obtained from the image to be identified.

[0017] In some embodiments, when the knife switch type is not a three-column horizontal rotary type, the knife switch includes two knife arms; the detection of the positional relationship between the knife arms in the target image of the knife switch and the preset knife switch structural component corresponding to the knife arms includes detecting the positional relationship between one of the two knife arms in the target image of the knife switch and the other knife arm.

[0018] In some embodiments, detecting the positional relationship between one of the two blade arms in the target image of the knife gate includes detecting the included angle between the two blade arms in the target image of the knife gate.

[0019] In some embodiments, when the knife switch type is a three-column horizontal rotary type, the knife switch includes a knife arm and two connecting components; the detection of the positional relationship between the knife arm in the target image of the knife switch and the preset knife switch structural component corresponding to the knife arm includes detecting the positional relationship between the knife arm in the target image of the knife switch and the two connecting components.

[0020] In some embodiments, detecting the positional relationship between the knife arm in the knife switch target image and the two connecting components includes detecting the included angle between the knife arm in the knife switch target image and the connecting line of the two connecting components.

[0021] In some embodiments, when the disconnector is not a three-column horizontal rotary type, determining the state of the disconnector based on its basic open / closed state, the disconnector type, and the positional relationship includes:

[0022] If the basic open / closed state of the disconnector is the closed state, and the condition γ∈[180°-α,180°+α] is satisfied, then the state of the disconnector is closed in place;

[0023] If the basic open / closed state of the disconnector is the closed state and does not satisfy the condition γ∈[180°-α,180°+α], then the state of the disconnector is an abnormal state.

[0024] If the basic open / closed state of the disconnector is open, and the condition γ∈[0°,β] is satisfied, then the state of the disconnector is open in place;

[0025] If the basic open / closed state of the disconnector is open and does not satisfy the condition γ∈[0°,β], then the state of the disconnector is abnormal.

[0026] Wherein, γ is the included angle between the two knife arms in the target image of the knife switch, α is the closing threshold, α∈[0°, 90°], and β is the opening threshold, β∈[0°, 90°].

[0027] In some embodiments, determining the state of the disconnector based on its basic open / closed state, the disconnector type, and its positional relationship includes:

[0028] If the basic open / closed state of the disconnector is the closed state, and the condition γ∈[180°-α,180°+α] is satisfied, then the state of the disconnector is closed in place;

[0029] If the basic open / closed state of the disconnector is the closed state and does not satisfy the condition γ∈[180°-α,180°+α], then the state of the disconnector is an abnormal state.

[0030] If the basic open / closed state of the disconnector is open, and the condition γ∈[β, 360°] is satisfied, then the state of the disconnector is open in place;

[0031] If the basic open / closed state of the disconnector is open and does not satisfy the condition γ∈[β, 360°], then the state of the disconnector is abnormal.

[0032] Wherein, γ is the angle between the knife arm in the target image of the knife switch and the connecting line of the two connecting parts, α is the closing threshold, α∈[0°, 90°], and β is the opening threshold, β∈[0°, 90°].

[0033] In some embodiments, the method further includes,

[0034] According to a preset semantic segmentation model, the target image of the disconnect switch is segmented to obtain the knife arm mask image of the disconnect switch; the knife arm mask image includes the knife arm target region and knife arm type information; the knife arm type information is used to distinguish the two knife arms in the target image of the disconnect switch.

[0035] Calculate the angles between the two cutter arms and the preset axis based on the cutter arm mask diagram, and calculate the angle between the two cutter arms based on the angles between the two cutter arms and the preset axis.

[0036] In some embodiments, calculating the angles between the two cutter arms and a preset axis based on the cutter arm mask diagram includes:

[0037] Based on the target area of ​​the cutting arm and the cutting arm type information, the corresponding first cutting arm target areas of the two cutting arms are obtained by using a preset algorithm, and the bounding rectangle with the smallest area of ​​the first cutting arm target area is calculated.

[0038] Calculate the angle between the line containing the preset side of the circumscribed rectangle and the preset axis.

[0039] Secondly, this embodiment provides a disconnector status identification device, the device including a detection module, used to detect an image to be identified containing a disconnector according to a preset first target detection model, and obtain the disconnector type and basic open / closed state of the disconnector in the image to be identified;

[0040] The processing module is used to determine the positional relationship between the knife arm and the knife switch structural component of the knife switch based on the image to be identified; the knife switch structural component is pre-set for the knife arm;

[0041] The identification module is used to determine the state of the disconnector based on the basic open / closed state of the disconnector, the disconnector type, and the positional relationship.

[0042] Thirdly, this embodiment provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the switch status identification method described in the first aspect.

[0043] Fourthly, this embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the disconnector status identification method described in the first aspect.

[0044] Compared with related technologies, the disconnector status identification method, device, electronic device and storage medium provided in this embodiment determine the final status of the disconnector by utilizing the type of disconnector, the initial closing state of the disconnector and the positional relationship, thereby improving the accuracy of disconnector status identification.

[0045] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0046] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0047] Figure 1 This is a hardware structure block diagram of the terminal executing the disconnector status identification method of this embodiment;

[0048] Figure 2 This is a flowchart of a disconnector status identification method according to this embodiment;

[0049] Figure 3 This is a flowchart of another disconnector status identification method in this embodiment;

[0050] Figure 4 This is a flowchart of another disconnector status identification method in this embodiment;

[0051] Figure 5 This is a flowchart of the disconnector status identification method according to a preferred embodiment;

[0052] Figure 6 This is a schematic diagram of the knife arm mask in this preferred embodiment;

[0053] Figure 7 This is a schematic diagram of the mask of the detection arm circumscribed rectangle in this preferred embodiment;

[0054] Figure 8 This is a structural block diagram of a three-column horizontal rotary disconnector according to a preferred embodiment;

[0055] Figure 9 This is a structural block diagram of a disconnector status identification device according to this embodiment. Detailed Implementation

[0056] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0057] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning as understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these,” used in this application, do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” and “coupled,” used in this application, are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. The term “multiple” used in this application refers to two or more. The "and / or" operator describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: A alone, A and B simultaneously, and B alone. Typically, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," and "third," etc., used in this application are merely for distinguishing similar objects and do not represent a specific ordering of the objects.

[0058] The method embodiments provided in this example can be executed on a terminal, computer, server, or similar computing device. The following description uses running on a terminal as an example. Figure 1 This is a hardware structure block diagram of the terminal executing the disconnector status identification method of this embodiment. For example... Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 and a memory 104 for storing data are also included. The processor 102 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown are illustrated.

[0059] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the switch status identification method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0060] The transmission device 106 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0061] This embodiment provides a method for identifying the status of a disconnect switch. Figure 2 This is a flowchart of a disconnector status identification method according to this embodiment, such as... Figure 2 As shown, the process includes the following steps:

[0062] Step S210: Based on the preset first target detection model, the image to be identified, containing the disconnector, is detected to obtain the disconnector type and basic open / closed state of the disconnector in the image to be identified. Specifically, after obtaining the image to be identified, the image contains the main body of the disconnector whose current state needs to be identified. The image to be identified is detected according to the preset first target detection model. For example, the YOLOv5S target detection model can be used to detect the image to be identified. The image to be identified is input into the pre-trained YOLOv5S target detection model, and the model outputs the disconnector type and basic open / closed state. The disconnector type includes double-column horizontal rotary disconnector, triple-column horizontal rotary disconnector, single-arm horizontal telescopic disconnector, single-arm vertical telescopic disconnector, and double-arm vertical telescopic disconnector, etc., which are not specifically limited here; the basic open / closed state of the disconnector is the open / closed state of the disconnector initially identified according to the first target detection model, where the basic open / closed state of the disconnector is either closed or open.

[0063] Step S220: Based on the image to be identified, determine the positional relationship between the knife arm and the knife gate structural component; the knife gate structural component is pre-set for the knife arm.

[0064] Specifically, the knife switch structural components, such as the knife arm and connecting parts, are detected from the image to be identified. Here, the knife switch structural components corresponding to the knife arm can be preset based on the knife switch type detected in step S210. For different types of knife switches, the positional relationship between the knife arm and the preset knife switch structural component corresponding to the knife arm can characterize the relative position between the knife arm and the knife switch structural component, and the state of the knife switch can be determined more accurately based on the relative position of the knife arm. For example, when the knife switch type is not a three-column horizontal rotary type, the knife switch includes two knife arms. Detecting the positional relationship between the knife arm in the image to be identified and the preset knife switch structural component corresponding to the knife arm can be done by detecting the positional relationship between one of the two knife arms in the image to be identified and the other knife arm. For a non-three-column horizontal rotary knife switch, when the basic open / closed state of the knife switch is closed, if the two knife arms are basically parallel, the state of the knife switch is determined to be closed; when the basic open / closed state of the knife switch is open, if the two knife arms are basically parallel, the state of the knife switch is determined to be open; otherwise, the state of the knife switch is determined to be abnormal. When the disconnect switch is a three-post horizontal rotary type, it includes a knife arm and two connecting parts. Detecting the positional relationship between the knife arm in the image to be identified and the preset knife switch structural components corresponding to the knife arm can be done by detecting the positional relationship between the knife arm and the two connecting parts in the image to be identified. For a three-post horizontal rotary disconnect switch, when the basic open / closed state of the disconnect switch is closed, if the line connecting the knife arm and the two connecting parts is basically parallel, the disconnect switch is determined to be in the closed position. When the basic open / closed state of the disconnect switch is open, if the maximum distance between the knife arm and the line connecting the two connecting parts is greater than a preset distance or the angle between the knife arm and the line connecting the two connecting parts is greater than a preset angle, the disconnect switch is determined to be in the open position. In other cases, the disconnect switch is determined to be in an abnormal state.

[0065] Step S230: Determine the status of the disconnector based on the disconnector's foundation open / closed state, disconnector type, and positional relationship.

[0066] Specifically, the state of the disconnector is determined based on the basic open / closed state and disconnector type detected in step S210 and the positional relationship determined in step S220. Here, the disconnector state includes closed, open, and abnormal states. This embodiment uses a preset target detection model to detect the image to be identified, obtaining the disconnector type and initial closed state. Based on the image to be identified, the positional relationship between the disconnector arm and the disconnector structural components is detected. Based on the disconnector type, initial closed state, and positional relationship, the final state of the disconnector is determined. Using the disconnector type, initial closed state, and positional relationship to determine the final state of the disconnector improves the accuracy of disconnector state identification.

[0067] Furthermore, in some embodiments, based on the above step S220, the positional relationship between the knife arm and the knife switch structural component is determined based on the image to be identified, specifically including the following steps:

[0068] Step S222: Obtain the target image of the knife switch containing the knife switch from the image to be identified; the area ratio of the knife switch in the target image of the knife switch is greater than the area ratio threshold.

[0069] Specifically, the image to be identified is detected to obtain a target image containing the disconnect switch. For example, after detecting the main body of the disconnect switch, the smallest circumscribed polygon corresponding to the main body area is taken as the boundary of the disconnect switch target. Here, the circumscribed polygon can be a circumscribed rectangle. The image area bounded by the disconnect switch target boundary is determined as the disconnect switch target image. Alternatively, the image area bounded by the disconnect switch target boundary after expanding it to a certain extent can also be determined as the target image. The area proportion of the disconnect switch in the disconnect switch target image is greater than the area proportion threshold to facilitate subsequent detection of disconnect switch structural components.

[0070] Step S224: Detect the knife arm and knife gate structural components in the target image of the knife gate, and determine the positional relationship between the knife arm and the preset knife gate structural components based on the detection results.

[0071] Specifically, the knife switch structural components, such as knife arms and connectors, in the target image of the knife switch are detected. Here, a knife switch structural component corresponding to the knife arm can be preset based on the knife switch type detected in step S210. For different types of knife switches, the positional relationship between the knife arm and the preset knife switch structural component corresponding to the knife arm can characterize the relative position between the knife arm and the knife switch structural component, and the state of the knife switch can be determined more accurately based on the relative position of the knife arm. Further, in some embodiments, when detecting the image to be identified containing the knife switch based on the above step S210 to obtain the knife switch type and basic open / closed state of the knife switch in the image to be identified, the following steps are also included:

[0072] Step S212: Based on the preset first target detection model, the image to be identified, which contains the knife switch, is detected to obtain the knife switch target boundary in the image to be identified.

[0073] Specifically, based on a preset first target detection model, the image to be identified is detected to obtain the target boundary of the disconnect switch, the type of disconnect switch, and the basic open / closed state of the disconnect switch. For example, the YOLOv5S target detection model can be used to detect the image to be identified. The image to be identified is input into the pre-trained YOLOv5S target detection model, and the model outputs the target boundary of the disconnect switch, the type of disconnect switch, and the basic open / closed state of the disconnect switch. Among them, the target boundary of the disconnect switch is the boundary of the area where the disconnect switch body is located in the image to be identified; the disconnect switch type includes double-column horizontal rotary disconnect switch, three-column horizontal rotary disconnect switch, single-arm horizontal telescopic disconnect switch, single-arm vertical telescopic disconnect switch, and double-arm vertical telescopic disconnect switch, etc., which are not specifically limited here; the basic open / closed state of the disconnect switch is the open / closed state of the disconnect switch initially identified according to the first target detection model, where the basic open / closed state of the disconnect switch is either closed or open.

[0074] Based on obtaining the boundary of the disconnector target, and according to the above step S222, the disconnector target image containing the disconnector is obtained from the image to be identified, specifically including the following steps:

[0075] Step S223: Based on the target boundary of the disconnector, obtain the target image of the disconnector containing the disconnector from the image to be identified.

[0076] Specifically, based on the target boundary of the disconnect switch obtained in step S212, the image to be identified is processed to obtain the target image of the disconnect switch. The image area defined by the target boundary of the disconnect switch can be determined as the target image of the disconnect switch, or the image area defined by expanding the target boundary of the disconnect switch to a certain extent can be determined as the target image.

[0077] In some embodiments, detecting the positional relationship between the knife arm in the knife switch target image and a preset knife switch structure corresponding to the knife arm can be achieved by detecting the distance between the knife arm in the knife switch target image and the preset knife switch structure corresponding to the knife arm.

[0078] In some embodiments, detecting the positional relationship between the knife arm in the knife switch target image and the preset knife switch structure corresponding to the knife arm can be done by detecting the included angle between the knife arm in the knife switch target image and the preset knife switch structure corresponding to the knife arm, or by detecting the included angle between the connecting line between the knife arm in the knife switch target image and the preset knife switch structure corresponding to the knife arm.

[0079] This embodiment also provides a method for identifying the status of a disconnect switch. Figure 3 This is a flowchart of another disconnector status identification method in this embodiment, such as... Figure 3 As shown, the process includes the following steps:

[0080] Step S310: Detect the image to be identified according to the preset first target detection model to obtain the target boundary, type, and basic open / closed state of the switch in the image to be identified.

[0081] Step S320: Based on the target boundary of the disconnector, obtain a target image of the disconnector containing only the disconnector from the image to be identified.

[0082] Step S330: When the switch type is not a three-column horizontal rotation type, detect the included angle between the two blade arms in the switch target image.

[0083] Specifically, when the disconnector type is not a three-column horizontal rotary type, the disconnector includes two blade arms. Using a pre-defined semantic segmentation model, the included angle between the two blade arms in the target image of the disconnector is detected.

[0084] Step S340: Determine the status of the disconnector based on the disconnector's foundation open / closed state, disconnector type, and included angle.

[0085] Specifically, if the disconnector base is in a closed state and satisfies the condition γ∈[180°-α,180°+α], then the disconnector is judged to be in a closed state; if the disconnector base is in a closed state but does not satisfy the condition γ∈[180°-α,180°+α], then the disconnector is judged to be in an abnormal state; if the disconnector base is in an open state and satisfies the condition γ∈[0°,β], then the disconnector is judged to be in a closed state; if the disconnector base is in an open state and does not satisfy the condition γ∈[0°,β], then the disconnector is judged to be in an abnormal state. Here, γ is the angle between the two disconnector arms in the disconnector target image, α is the closing threshold (α∈[0°, 90°]), and β is the opening threshold (β∈[0°, 90°]).

[0086] This embodiment improves the accuracy of disconnector status detection by detecting the included angle between the two cutter arms and determining the final state of the disconnector based on the included angle.

[0087] Furthermore, in some embodiments, based on the above step S330, detecting the included angle between the two blade arms in the target image of the blade switch specifically includes the following steps:

[0088] Step S332: Based on a preset semantic segmentation model, the target image of the disconnector is segmented to obtain a mask image of the disconnector's blade arms. The blade arm mask image includes the target region of the blade arm and the blade arm type information. The blade arm type information is used to distinguish the two blade arms in the target image of the disconnector.

[0089] Specifically, the target image of the disconnector is input into a pre-trained semantic segmentation model. The model outputs a mask image of the disconnector's blade arms, which includes the target region of the blade arm and information about the blade arm type. Different blade arm types of disconnectors include left blade arm, right blade arm, upper blade arm, and lower blade arm. For example, the blade arm type information can be the background information of the target region of different blade arm types in the blade arm mask image.

[0090] In some embodiments, a lightweight HrNet semantic segmentation model can be used to segment the target image of the disconnector to obtain the mask image of the disconnector arm. The backbone network of this lightweight HrNet semantic segmentation model has 18 input channels in the second to fourth stages, and 2 BASIC blocks in the first to fourth stages. The lightweight HrNet semantic segmentation model reduces model parameters by decreasing the number of input channels and blocks, thereby reducing the computational cost.

[0091] Step S334: Calculate the angles between the two cutter arms and the preset axis according to the cutter arm mask diagram, and calculate the angle between the two cutter arms based on the angles between the two cutter arms and the preset axis.

[0092] Specifically, based on the target area and type information of the cutting arm obtained in step S332, the first target area of ​​each of the two cutting arms is obtained using a preset algorithm. The bounding rectangle with the smallest area of ​​the first target area is calculated, and the angle between the line containing the preset side of the bounding rectangle and the preset axis is calculated. The maximum connected component algorithm can be used to find the largest area in the target area of ​​each type of cutting arm, and the bounding rectangle with the smallest area of ​​the largest area is calculated.

[0093] This embodiment utilizes a semantic segmentation model to define the two blade arms of the switch as different types. The model outputs blade arm type information and target area of ​​the blade arm, thus solving the problem that the angle cannot be calculated due to the overlap of the two blade arms in the image.

[0094] This embodiment also provides a method for identifying the status of a disconnect switch. Figure 4 This is a flowchart of another disconnector status identification method in this embodiment, such as... Figure 4 As shown, the process includes the following steps:

[0095] Step S410: Detect the image to be identified according to the preset first target detection model to obtain the target boundary, type, and basic open / closed state of the switch in the image to be identified.

[0096] Step S420: Based on the target boundary of the disconnector, obtain a target image of the disconnector containing only the disconnector from the image to be identified.

[0097] Step S430: When the switch type is a three-column horizontal rotary switch, detect the angle between the switch arm and the connecting line of the two connecting parts in the switch target image.

[0098] Specifically, a pre-trained second target detection model is used to detect the blade arm and two connecting components in the target image of the knife gate. Based on the detected blade arm and connecting components, the angle between the connecting lines of the blade arm and the two connecting components is calculated. The second target detection model here can be the same as or different from the first target detection model; no specific limitation is made here.

[0099] Step S440: Determine the status of the disconnector based on the disconnector's foundation open / closed state, disconnector type, and included angle.

[0100] Specifically, if the disconnector base is in the closed state and satisfies the condition γ∈[180°-α,180°+α], then the disconnector is in the closed position. If the disconnector base is in the closed state and does not satisfy the condition γ∈[180°-α,180°+α], then the disconnector is in an abnormal state. If the disconnector base is in the open state and satisfies the condition γ∈[β,360°], then the disconnector is in the open position. If the disconnector base is in the open state and does not satisfy the condition γ∈[β,360°], then the disconnector is in an abnormal state. Wherein, γ is the angle between the disconnector arm and the connecting line of the two connecting parts in the disconnector target image, α is the closing position threshold, α∈[0°,90°], and β is the opening position threshold, β∈[0°,90°].

[0101] This embodiment detects the angle between the knife arm and the connecting lines of the two connecting components, and determines the final state of the knife switch based on the angle between the knife arm and the connecting lines of the two connecting components, thus further improving the accuracy of knife switch state detection.

[0102] The present embodiment will now be described and illustrated through preferred embodiments.

[0103] Figure 5 This is a flowchart of the disconnector status identification method according to a preferred embodiment. Figure 5 As shown, the disconnector status identification method includes the following steps:

[0104] Step S501: Configure the closing threshold α and the opening threshold β.

[0105] Specifically, α is the closing threshold, α∈[0°, 90°], β is the opening threshold, β∈[0°, 90°], α is preferably 10°, and β is preferably 20°.

[0106] Step S502: Detect the main body range of the disconnector, the disconnector type, and the open / closed status of the disconnector foundation. If the disconnector type is a three-column horizontal rotary type, proceed to step S506; otherwise, proceed to step S503.

[0107] First, data is collected by cameras, capturing scene images of the substation containing disconnector devices. The collected data is divided into training and testing sets, and the main body range, type, and open / closed status of the disconnectors are labeled. Then, the YOLOv5S object detection model is trained using the training set data, and tested using the testing set data. After model training and testing, the image to be detected is used as input to the model, and the model outputs the main body range, type, and open / closed status of the disconnector. The image to be detected is then cropped based on the main body range of the disconnector to obtain the target image of the disconnector.

[0108] Step S503: Segment the knife arm in the knife gate target image.

[0109] The acquired images need to be labeled with the range and category of the cutter arms, including left cutter arm, right cutter arm, upper cutter arm, and lower cutter arm. Training and testing sets are then created. The lightweight HRNet semantic segmentation model is trained using the training set data, and tested using the test set data. Finally, the model segments the cutter arm range by inputting the test set data. After model training and testing, the target image of the cutter gate obtained in step S502 is used as the model input, and the model outputs a cutter arm mask image, which includes the cutter arm region and cutter arm category. For example, for a cutter gate with left and right cutter arms, its cutter arm mask image is as follows. Figure 6 As shown, Figure 6 This is a schematic diagram of the tool arm mask according to a preferred embodiment. The tool arm mask includes a left tool arm target area 601, a right tool arm target area 602, and tool arm type information. The left tool arm target area 601 is the area where the left tool arm is located, and the right tool arm target area 602 is the area where the right tool arm is located. Figure 6 In the cutter arm mask image, the backgrounds of the left and right cutter arms are different to distinguish different types of cutter arms.

[0110] Step S504: Detect the profile of the cutter arm.

[0111] Using the cutter arm mask image obtained in step S503, the maximum connected component algorithm is used to find the region with the largest area in each type of cutter arm and calculate the bounding rectangle with the smallest area of ​​that region. Figure 7 This is a schematic diagram of the cutter arm mask for the detection cutter arm's circumscribed rectangle, according to a preferred embodiment. Figure 7As shown, the maximum connected component algorithm is used to find the maximum region 701 of the left cutting arm target region 601, and the maximum connected component algorithm is used to find the maximum region 702 of the right cutting arm target region 602. The minimum area bounding rectangle 703 of the left cutting arm's maximum region 701 is calculated, and the minimum area bounding rectangle 704 of the right cutting arm's maximum region 702 is calculated. Specifically, the minimum area bounding rectangle 703 of the left cutting arm is the rectangle with the smallest area within the bounding rectangle of the left cutting arm's maximum region 701, and the minimum area bounding rectangle 704 of the right cutting arm's maximum region 702 is the rectangle with the smallest area within the bounding rectangle of the right cutting arm's maximum region 702. For example, when both the left cutter arm target region 601 and the right cutter arm target region 602 have only one connected region, the largest region 701 of the left cutter arm is the left cutter arm target region 601, and the largest region 702 of the right cutter arm is the right cutter arm target region 602. When there is occlusion in the cutter arm region of the cutter gate target image, a cutter arm may be divided into multiple regions, and the left cutter arm target region 601 or the right cutter arm target region 602 may have multiple connected regions. In this case, the region with the largest area in the left cutter arm target region 601 is taken as the largest region 701 of the left cutter arm, and the region with the largest area in the right cutter arm target region 602 is taken as the largest region 702 of the right cutter arm.

[0112] Step S505: Calculate the cutter arm angle.

[0113] Based on the circumscribed rectangle of the cutter arm obtained in step S504, calculate the angles between the lines containing the long sides of the two cutter arms and the horizontal direction, denoted as a and b respectively. First, calculate the slope k based on the two endpoints of the cutter arm lines: θ = arctan(k) * 180 / π. If θ < 0, then θ = 180 + θ.

[0114] Step S506: Inspect the rotating parts and connecting parts.

[0115] When the type of switch detected in step S502 is a three-post horizontal rotary switch, the two blades of the three-post horizontal rotary switch are integrated and the included angle is always 0. Therefore, it is necessary to detect the rotating parts and the connecting parts. Figure 8 This is a structural block diagram of a three-column horizontal rotary disconnector according to a preferred embodiment of the present invention, as shown below. Figure 8As shown, the knife switch includes a knife arm 801, a rotating component 802, a first connecting component 803, and a second connecting component 804. The angle between the knife arm and the connecting component is calculated. The knife arm, rotating component, and connecting component are detected in the knife switch target image obtained in step S502 using the YOLOv5S target detection model. Specifically, the knife switch target image is divided into a training set and a test set, and the knife arm, rotating component, and connecting component are labeled. Subsequently, the YOLOv5S target detection model is trained using the data from the training set, and the model is tested using the data from the test set. After the model training and testing are completed, the knife switch target image is used as the model input, and the model outputs the knife arm, rotating component, and connecting component. The angle c between the knife arm 801 and the horizontal direction, and the angle d between the connecting line of the first connecting component 803 and the second connecting component 804 and the horizontal direction are calculated.

[0116] Step S507: Calculate the included angle γ.

[0117] Based on the angles of the two switch arms obtained in step S505 or the angle obtained in step S506, calculate the included angle γ1, where γ1 = |ab| or γ1 = |cd|. Based on the switch arm status obtained in step S502, if the switch base is in the closed state and γ1 is less than 90°, then γ = 180° - γ1; if the switch base is in the open state and γ1 is greater than 90°, then γ = 180° - γ1; otherwise, γ = γ1.

[0118] Step S508: Identify the status of the disconnector.

[0119] If the disconnector type is not a three-column horizontal rotary type, then,

[0120] (1) The basic opening and closing status of the disconnector detected in step S502 is closed and 180°-α<=γ<=180°+α, and the disconnector status is closed in place;

[0121] (2) The disconnector status detected in step S502 is open and 0°<=γ<=β, the disconnector status is in the open position;

[0122] (3) In other cases, the switch status is abnormal.

[0123] If the disconnect switch type is a three-post horizontal rotary type, then,

[0124] (1) The basic opening and closing status of the disconnector detected in step S502 is closed and 180°-α<=γ<=180°+α, and the disconnector status is closed in place;

[0125] (2) The basic open / closed state of the disconnector detected in step S502 is open and γ>=β, and the disconnector state is open in place;

[0126] (3) In other cases, the switch status is abnormal.

[0127] This embodiment utilizes an object detection model to obtain the type and open / closed status of the disconnector, providing accurate prior information for calculating the included angle of different types of disconnectors and improving versatility. A semantic segmentation model is used to define two disconnector arms as different categories, solving the problem of being unable to calculate the included angle when two disconnector arms overlap. Traditional image processing methods are greatly affected by changes in lighting and have poor robustness. Disconnector status recognition based on a deep learning model using object detection and semantic segmentation improves the accuracy of disconnector status recognition.

[0128] This embodiment also provides a disconnector status identification device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. The terms "module," "unit," "subunit," etc., used below refer to combinations of software and / or hardware that implement a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0129] Figure 9 This is a structural block diagram of a disconnector status identification device according to this embodiment, as shown below. Figure 9 As shown, the device includes:

[0130] The detection module 910 is used to detect the image to be identified, which contains the disconnect switch, according to the preset first target detection model, and to obtain the disconnect switch type and basic open / closed state of the disconnect switch in the image to be identified.

[0131] Processing module 920 is used to determine the positional relationship between the knife arm and the knife switch structural components of the knife switch based on the image to be identified; the knife switch structural components are pre-set for the knife arm;

[0132] The identification module 930 is used to determine the status of the disconnector based on the basic open / closed state of the disconnector, the disconnector type, and the positional relationship.

[0133] This embodiment also provides an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0134] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0135] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0136] S1, based on the preset first target detection model, detect the image to be identified to obtain the target boundary, type and basic open / closed state of the switch in the image to be identified;

[0137] S2, Based on the target boundary of the disconnector, obtain a target image of the disconnector containing only the disconnector from the image to be identified;

[0138] S3, detect the positional relationship between the knife arm in the knife switch target image and the preset knife switch structural component corresponding to the knife arm;

[0139] S4. Determine the status of the disconnector based on the open / closed state of the disconnector foundation, the type of disconnector, and its positional relationship.

[0140] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated in this embodiment.

[0141] Furthermore, in conjunction with the disconnector status identification method provided in the above embodiments, this embodiment can also provide a storage medium for implementation. The storage medium stores a computer program; when executed by a processor, the computer program implements the steps of any of the disconnector status identification methods in the above embodiments.

[0142] It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0143] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0144] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or independent of other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0145] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A method for identifying the status of a disconnect switch, characterized in that, The method includes, According to the preset first target detection model, the image to be identified containing the disconnect switch is detected to obtain the disconnect switch type and basic open / closed state of the disconnect switch in the image to be identified; Based on the image to be identified, determine the positional relationship between the knife arm and the knife switch structural components of the knife switch; The knife gate structure is pre-configured for the knife arm; The state of the disconnector is determined based on the basic open / closed state of the disconnector, the type of disconnector, and the positional relationship. When the disconnector type is not a three-column horizontal rotary type, the disconnector includes two blade arms; determining the positional relationship between the blade arms and the disconnector structural components based on the image to be identified includes: According to a preset semantic segmentation model, the target image of the disconnector containing the disconnector in the image to be identified is segmented to obtain the knife arm mask map of the disconnector; the knife arm mask map includes the knife arm target area and knife arm type information; the knife arm type information is used to distinguish the two knife arms in the target image of the disconnector. Based on the target region of the cutting arm and the cutting arm type information, the first target region of each cutting arm is obtained by using the maximum connected component algorithm, and the bounding rectangle with the smallest area of ​​the first target region of the cutting arm is calculated. Calculate the angle between the line containing the preset side of the circumscribed rectangle and the preset axis, and calculate the angle between the two cutter arms based on the angle between the two cutter arms and the preset axis.

2. The disconnector status identification method according to claim 1, characterized in that, The step of determining the positional relationship between the knife arm and the knife switch structural components based on the image to be identified includes: From the image to be identified, a target image of the disconnect switch containing the disconnect switch is obtained; the area of ​​the disconnect switch in the target image of the disconnect switch is greater than the area ratio threshold. The step of determining the positional relationship between the knife arm and the knife switch structural components based on the image to be identified includes: The knife arm and the knife gate structure in the target image of the knife gate are detected, and the positional relationship between the knife arm and the preset knife gate structure is determined based on the detection results.

3. The disconnector status identification method according to claim 2, characterized in that, When detecting an image containing a disconnector to obtain the disconnector type and basic open / closed state of the disconnector in the image, the process further includes: According to the preset first target detection model, the image to be identified, which contains the knife switch, is detected to obtain the knife switch target boundary in the image to be identified; Obtaining the target image of the disconnector containing the disconnector from the image to be identified includes: Based on the target boundary of the disconnector, the target image of the disconnector containing the disconnector is obtained from the image to be identified.

4. The disconnector status identification method according to claim 1, characterized in that, When the type of the knife switch is a three-column horizontal rotary type, the knife switch includes a knife arm and two connecting components; the detection of the positional relationship between the knife arm in the target image of the knife switch and the preset knife switch structural component corresponding to the knife arm includes detecting the positional relationship between the knife arm in the target image of the knife switch and the two connecting components.

5. The disconnector status identification method according to claim 4, characterized in that, The step of detecting the positional relationship between the knife arm in the target image of the knife switch and the two connecting components includes detecting the angle between the knife arm in the target image of the knife switch and the connecting line of the two connecting components.

6. The disconnector status identification method according to claim 1, characterized in that, When the disconnector is a non-three-column horizontal rotary type, determining the state of the disconnector based on its foundation open / closed state, the disconnector type, and the positional relationship includes: If the basic open / closed state of the disconnector is the closed state, and the condition γ∈[180°-α,180°+α] is satisfied, then the state of the disconnector is closed in place; If the basic open / closed state of the disconnector is the closed state and does not satisfy the condition γ∈[180°-α,180°+α], then the state of the disconnector is an abnormal state. If the basic open / closed state of the disconnector is open, and the condition γ∈[0°,β] is satisfied, then the state of the disconnector is open in place; If the basic open / closed state of the disconnector is open and does not satisfy the condition γ∈[0°,β], then the state of the disconnector is abnormal. Wherein, γ is the included angle between the two knife arms in the target image of the knife switch, α is the closing threshold, α∈[0°, 90°], and β is the opening threshold, β∈[0°, 90°].

7. The disconnector status identification method according to claim 5, characterized in that, Determining the state of the disconnector based on its basic open / closed state, the disconnector type, and its positional relationship includes: If the basic open / closed state of the disconnector is the closed state, and the condition γ∈[180°-α,180°+α] is satisfied, then the state of the disconnector is closed in place; If the basic open / closed state of the disconnector is the closed state and does not satisfy the condition γ∈[180°-α,180°+α], then the state of the disconnector is an abnormal state. If the basic open / closed state of the disconnector is open, and the condition γ∈[β, 360°] is satisfied, then the state of the disconnector is open in place; If the basic open / closed state of the disconnector is open and does not satisfy the condition γ∈[β, 360°], then the state of the disconnector is abnormal. Wherein, γ is the angle between the knife arm in the target image of the knife switch and the connecting line of the two connecting parts, α is the closing threshold, α∈[0°, 90°], and β is the opening threshold, β∈[0°, 90°].

8. A disconnect switch status identification device, characterized in that, The device includes, The detection module is used to detect the image to be identified, which contains a disconnector, according to a preset first target detection model, and to obtain the disconnector type and basic open / closed state of the disconnector in the image to be identified. The processing module is used to determine the positional relationship between the knife arm and the knife switch structural components of the knife switch based on the image to be identified; The knife gate structure is pre-configured for the knife arm; The identification module is used to determine the state of the disconnector based on the basic open / closed state of the disconnector, the disconnector type, and the positional relationship. When the disconnector type is not a three-column horizontal rotary type, the disconnector includes two blade arms; determining the positional relationship between the blade arms and the disconnector structural components includes: According to a preset semantic segmentation model, the target image of the disconnector containing the disconnector in the image to be identified is segmented to obtain the knife arm mask map of the disconnector; the knife arm mask map includes the knife arm target area and knife arm type information; the knife arm type information is used to distinguish the two knife arms in the target image of the disconnector. Based on the target region of the cutting arm and the cutting arm type information, the first target region of each cutting arm is obtained by using the maximum connected component algorithm, and the bounding rectangle with the smallest area of ​​the first target region of the cutting arm is calculated. Calculate the angle between the line containing the preset side of the circumscribed rectangle and the preset axis, and calculate the angle between the two cutter arms based on the angle between the two cutter arms and the preset axis.

9. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the disconnector status identification method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the disconnector status identification method according to any one of claims 1 to 7.

Images