Grounding knife switch state determination method and device, electronic equipment and storage medium

Through the adversarial network model training of the grounding knife switch image sample set, a state determination model is generated, which solves the problem of low efficiency and safety hazards in the judgment of the grounding knife switch split and closing state in the substation, and realizes efficient and accurate automated judgment, reducing human errors risks.

CN119068227APending Publication Date: 2024-12-03GUANGDONG POWER GRID CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410847734.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-27
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In the prior art, the determination of the separation and closing state of the grounding knife switch in a substation depends on human eye observation, which is inefficient and has safety hazards, which may lead to equipment failure or maintenance personnel being in danger.

Method used

The adversarial network model is used to train the grounding knife switch image sample set, generate a state determination model, and determine the separation and combination state of the grounding knife switch through image recognition to improve judgment efficiency and accuracy, and avoid safety hazards caused by human eye observation.

Benefits of technology

It improves the efficiency and accuracy of the judgment of the split-closing state of the grounding knife switch, reduces the safety risks brought by human error, and provides reliable and automated judgment support for the operation, inspection and maintenance of the substation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068227A_ABST
    Figure CN119068227A_ABST
Patent Text Reader

Abstract

The invention discloses a grounding knife switch state determination method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a grounding knife-switch image sample set; training the adversarial network model according to the grounding knife-switch image sample set to generate a grounding knife-switch state determination model; and obtaining a target image of the target grounding knife-switch, inputting the target image to the grounding knife-switch state determination model, and determining a target on-off state of the target grounding knife-switch according to an output result of the grounding knife-switch state determination model. According to the technical scheme provided by the embodiment of the invention, the on-off state of the target grounding knife-switch is determined through the grounding knife-switch state determination model obtained by training the adversarial network model and the acquired target image, so that the efficiency and accuracy of judging the on-off state of the target grounding knife-switch are improved, and the accuracy of judging the on-off state of the target grounding knife-switch is improved. And potential safety hazards caused by human eye observation are avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method, device, electronic equipment and storage medium for determining a grounding switch state. Background Art

[0002] With the development of economy, the number of substations has also increased, which has greatly increased the workload of operation and maintenance. Switching operation is one of the important tasks of daily maintenance and operation of substations. This work requires accurate judgment of the opening and closing state of the grounding knife switch. Otherwise, when the equipment is powered on, if the grounding knife switch is not opened in place and the moving and static contacts are not completely separated, it will directly cause the failure of power supply with the grounding knife switch. During equipment maintenance, if the grounding knife switch is not closed in place, the maintenance equipment will not be effectively grounded, and will have induced voltage. When a sudden call comes, the current cannot be led to the ground, which brings great personal risks to the maintenance personnel. At present, the method of determining the opening and closing position of the grounding knife switch in most substations is still human visual observation. This method is inefficient and has certain safety hazards. Summary of the invention

[0003] The present invention provides a method, device, electronic device and storage medium for determining the state of a grounding switch, which improves the efficiency and accuracy of judging the open and close states of a target grounding switch and avoids potential safety hazards caused by human visual observation.

[0004] According to one aspect of the present invention, a method for determining a state of a grounding switch is provided, the method comprising:

[0005] Acquire a grounding switch image sample set;

[0006] The adversarial network model is trained according to the grounding switch image sample set to generate a grounding switch state determination model;

[0007] A target image of a target grounding switch is acquired, the target image is input into the grounding switch state determination model, and a target opening and closing state of the target grounding switch is determined according to an output result of the grounding switch state determination model.

[0008] According to another aspect of the present invention, a device for determining a state of a grounding switch is provided, the device comprising:

[0009] A grounding switch image sample set acquisition module is used to acquire a grounding switch image sample set;

[0010] A state determination model generation module is used to train the adversarial network model according to the grounding switch image sample set to generate a grounding switch state determination model;

[0011] The grounding switch state determination module is used to obtain a target image of a target grounding switch, input the target image into the grounding switch state determination model, and determine the target opening and closing state of the target grounding switch according to the output result of the grounding switch state determination model.

[0012] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the state of a grounding switch described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the state of a grounding switch described in any embodiment of the present invention when executed.

[0017] The technical solution of the embodiment of the present invention obtains a sample set of grounding switch images; trains an adversarial network model based on the sample set of grounding switch images to generate a grounding switch state determination model; obtains a target image of a target grounding switch, inputs the target image into the grounding switch state determination model, and determines the target open / close state of the target grounding switch based on the output result of the grounding switch state determination model. The technical solution of the embodiment of the present invention determines the open / close state of the target grounding switch by using the grounding switch state determination model obtained by training the adversarial network model and the target image obtained by acquisition, thereby improving the efficiency and accuracy of judging the open / close state of the target grounding switch and avoiding the safety hazards caused by human eye observation.

[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a flow chart of a method for determining a grounding switch state provided according to Embodiment 1 of the present invention;

[0021] Figure 2 is a flow chart of a method for determining a grounding switch state according to a second embodiment of the present invention;

[0022] Figure 3 2 is a schematic diagram of the structure of a device for determining the state of a grounding switch provided according to a third embodiment of the present invention;

[0023] Figure 4 It is a structural schematic diagram of an electronic device for implementing the method for determining the state of a grounding switch according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] Embodiment 1

[0027] Figure 1 A flow chart of a method for determining the state of a grounding switch is provided for the first embodiment of the present invention. This embodiment is applicable to the case of determining the on / off state of a grounding switch in a substation. The method can be executed by a device for determining the state of a grounding switch. The device for determining the state of a grounding switch can be implemented in the form of hardware and / or software. The device for determining the state of a grounding switch can be configured in an electronic device. Figure 1 As shown, the method includes:

[0028] S110, obtaining a grounding switch image sample set.

[0029] Among them, the grounding knife switch is an important power equipment, which is mainly used for switch operation on high-voltage transmission lines to ensure the safety and stable operation of the power system. The grounding knife switch is widely used in various high-voltage power equipment, such as transmission lines, substations, power switch equipment and various power stations. Among them, the grounding knife switch image sample set includes at least two grounding knife switch images with the open and closed states annotated, which are used to train the adversarial network model.

[0030] Optionally, obtaining a grounding switch image sample set includes: obtaining at least two grounding switch images from a grounding switch image library of the substation, and respectively determining the open and closed states of the grounding switches corresponding to the at least two grounding switch images; and marking the corresponding at least two grounding switch images based on the open and closed states to generate a grounding switch image sample set.

[0031] Among them, the opening and closing states of the grounding switch include the closed state and the open state, which can be represented by a 0-1 variable, such as 0 for the open state and 1 for the closed state.

[0032] In an embodiment of the present invention, after obtaining at least two grounding switch images from the grounding switch image library of the substation, the at least two grounding switch images can be preprocessed, such as grayscale processing, denoising, binarization processing, etc., to facilitate the subsequent model training process. After that, the opening and closing states of the grounding switches corresponding to the at least two grounding switch images are determined respectively, and the at least two grounding switch images are annotated based on the opening and closing states to generate a grounding switch image sample set.

[0033] It can be understood that a substation usually includes multiple grounding switches. When acquiring at least two images of the grounding switches, at least two historical images of each grounding switch can be acquired and arranged according to the serial numbers of the grounding switches in the substation to generate multiple groups of grounding switch image subsets. For each group of grounding switch image subsets, the historical images in the group of grounding switch image subsets are marked according to the opening and closing states of the grounding switches corresponding to the historical images in the group, to generate the grounding switch image sample subset shown below:

[0034] S={(x1, t1), (x2, t2)…(x n , t n )};

[0035] Among them, S represents the grounding switch image sample subset, x n Represents the historical image of the nth grounding switch in the substation, t nIt represents the on / off state of the grounding switch in the historical image of the nth grounding switch in the substation, and its value is 0 or 1. Afterwards, each grounding switch image sample subset is integrated to obtain a grounding switch image sample set.

[0036] S120: training the adversarial network model according to the grounding switch image sample set to generate a grounding switch state determination model.

[0037] Among them, the adversarial network model is a deep learning model, which consists of two competing neural networks, namely the generator and the discriminator. The generator is used to generate new samples similar to real data, which are called fuzzy data in the embodiment of the present invention, and the discriminator is used to determine whether the input sample is a real sample or a fake sample generated by the generator. Exemplarily, the adversarial network model can be expressed by the following formula:

[0038]

[0039] Among them, D is the discriminator part, G is the generator part, E represents the mathematical expectation, and p g(x) Indicates the probability distribution of the real data x, p z(z) represents the probability distribution of the input data z, D(x) represents the discriminator output of the real data, G(z) represents the fuzzy data generated by the generator based on the input data, It means that the above formula is maximized when training the discriminator and minimized when training the generator.

[0040] In an embodiment of the present invention, after obtaining a grounding switch image sample set, the adversarial network model can be trained according to the grounding switch image sample set to generate a grounding switch state determination model. During the training process, the generator and the discriminator are trained alternately, the discriminator is trained so that it can accurately distinguish between real images and fuzzy images, and the generator is trained so that the generated fuzzy images can be as close to the real images as possible. It should be noted that in an embodiment of the present invention, the grounding switch image sample set used also includes annotation information on the on-off state of the grounding switch in the grounding switch image. In the process of training the discriminator with the grounding switch image sample set, the discriminator can learn the relationship between the image features of the grounding switch image and the on-off state corresponding to the grounding switch image, so that the model obtained after the training is completed can accurately judge the on-off state of the grounding switch in the image according to the grounding switch image.

[0041] S130, obtaining a target image of a target grounding switch, inputting the target image into a grounding switch state determination model, and determining a target opening and closing state of the target grounding switch according to an output result of the grounding switch state determination model.

[0042] The target grounding switch is the grounding switch whose opening and closing states need to be determined. The number of target grounding switches may be one or more, such as multiple grounding switches in a substation.

[0043] In an embodiment of the present invention, when the number of target grounding switches is one, the target image of the target grounding switch can be directly obtained and input into the grounding switch state determination model, and the target open / close state of the target grounding switch can be determined based on the output result of the grounding switch state determination model; when the number of target grounding switches is multiple, multiple target images corresponding to multiple target grounding switches can be obtained, and they can be arranged in a certain order and input into the grounding switch state model, so that the grounding switch state model outputs the open / close state of multiple grounding switches in this order, so as to simultaneously complete the determination of the open / close state of multiple target grounding switches to avoid confusion.

[0044] The technical solution of the embodiment of the present invention obtains a grounding switch image sample set; trains an adversarial network model based on the grounding switch image sample set to generate a grounding switch state determination model; obtains a target image of a target grounding switch, inputs the target image into the grounding switch state determination model, and determines the target open / close state of the target grounding switch based on the output result of the grounding switch state determination model. The technical solution of the embodiment of the present invention determines the open / close state of the target grounding switch by using the grounding switch state determination model obtained by training the adversarial network model and the target image obtained by acquisition, thereby improving the efficiency and accuracy of judging the open / close state of the target grounding switch and avoiding the safety hazards caused by human eye observation.

[0045] Embodiment 2

[0046] Figure 2 This is a flow chart of a method for determining the state of a grounding switch provided in the second embodiment of the present invention. The embodiment of the present invention is optimized based on the above embodiment. For solutions not described in detail in the embodiment of the present invention, please refer to the above embodiment. Figure 2 As shown, the method includes:

[0047] S210, obtaining a grounding switch image sample set.

[0048] S220, inputting the grounding switch image sample set into the generator of the adversarial network model, and after the generator learns the grounding switch image sample set, generating a fuzzy image of each grounding switch image in the grounding switch image sample set.

[0049] The fuzzy image is an image generated by the generator based on the images in the grounding switch image sample set, which is similar to the images in the grounding switch image sample set. In the embodiment of the present invention, after the grounding switch image sample set is input into the generator, the generator will learn the features of the images in the grounding switch image sample set, and continuously adjust its own parameters through the back propagation algorithm to generate a fuzzy image similar to the images in the grounding switch image sample set.

[0050] S230, input the fuzzy image into the discriminator of the adversarial network model, guide the discriminator output through the open and close states annotated in the grounding switch image sample set, iteratively optimize the network parameters of the adversarial network model, and generate a grounding switch state determination model.

[0051] In an embodiment of the present invention, the fuzzy image is input into the discriminator of the adversarial network model, and the output of the discriminator is guided by the on-off state annotated in the grounding switch image sample set. In this process, the discriminator learns the relationship between the image features of the grounding switch image and the on-off state corresponding to the grounding switch image, so as to judge the on-off state of the grounding switch in the grounding switch image. It can be understood that the training of the adversarial network model is an iterative process. As the training proceeds, the network parameters of the adversarial network model itself are continuously iteratively optimized, and when the expected state is achieved, a grounding switch state determination model is generated.

[0052] Optionally, the network parameters of the adversarial network model are iteratively optimized to generate a grounding switch state determination model, including: in the process of iteratively optimizing the network parameters of the adversarial network model, determining the output error of the adversarial network model according to a target loss function, until the output error is less than a preset threshold, generating a grounding switch state determination model; wherein the target loss function includes an adversarial loss function, a conditional loss function, a perceptual loss function, and a gradient loss function.

[0053] Among them, the target loss function is used to determine the difference between the output value of the adversarial network model and the actual value. The target loss function includes the adversarial loss function, the conditional loss function, the perceptual loss function and the gradient loss function. Specifically, the target loss function can be expressed as the following formula:

[0054] L G =αL a +L b +λ1L c +λ2L d ;

[0055] Among them, L G represents the target loss function, L a represents the adversarial loss function, L b represents the conditional loss function, L crepresents the perceptual loss function, L d represents the gradient loss function, α represents the adversarial loss coefficient, and λ1 and λ2 represent the proportional weight coefficients. Among them, α, λ1 and λ2 can be determined according to the network parameters of the adversarial network.

[0056] Among them, the adversarial loss function L a It can be expressed as the following formula:

[0057]

[0058] Conditional loss function L b It can be expressed as the following formula:

[0059]

[0060] Perceptual loss function L c And the gradient loss function L d Can be composed to optimize the content loss function L y , optimize the content loss function L y It can be expressed as the following formula:

[0061]

[0062] Among them, D is the discriminator part, G is the generator part, E represents the mathematical expectation, and p g(x) Indicates the probability distribution of the real data x, p z(z) represents the probability distribution of the input data z, D(x) represents the discriminator output of the real data, G(z) represents the fuzzy data generated by the generator based on the input data, and I B represents the input image, C·H·W represents the number of channels and length and width of the gradient map, R represents the round of gradient calculation, d represents the gradient image, φ i,j represents the output feature map of the jth convolutional layer before the i-th maximum pooling layer after activation, W i,j H i,j Respectively represent the length and width of the feature map, I s represents the real image, G(I B ) represents the generated blurred image.

[0063] In an embodiment of the present invention, during the iterative optimization of the network parameters of the adversarial network model, the coefficients of each loss function in the target loss function can be determined according to the current network parameters, and then the target loss function at the current moment can be determined, and the output error of the current adversarial network model can be determined according to the target loss function. When judging whether the output error of the current adversarial network model is less than a preset threshold, if so, the iterative process is terminated and the current adversarial network model is used as the grounding switch state determination model. If not, the iterative process is repeated until the output error of the current adversarial network model is less than the preset threshold.

[0064] S240, obtaining a grounding switch image test set, and inputting the grounding switch image test set into a grounding switch state determination model.

[0065] In an embodiment of the present invention, at least two grounding switch images can be obtained from a grounding switch image library of a substation, and at least two grounding switch images are preprocessed, such as grayscale processing, denoising, binarization processing, etc., to generate a grounding switch image test set. Afterwards, the grounding switch image test set is input into a grounding switch state determination model to obtain a test result output by the grounding switch state determination model.

[0066] S250: Evaluate the accuracy of the grounding switch state determination model according to the test result output by the grounding switch state determination model.

[0067] In an embodiment of the present invention, the accuracy of the grounding switch state determination model can be evaluated according to the test results output by the grounding switch state determination model. Optionally, the ratio of the number of accurate output results to the total number of output results in the test results can be used as the accuracy of the grounding switch state determination model. Optionally, after determining the accuracy of the grounding switch state determination model, it also includes: if the accuracy of the grounding switch state determination model is lower than a preset accuracy threshold, optimizing and adjusting the grounding switch state determination model until the accuracy of the grounding switch state determination model is not lower than the preset accuracy threshold.

[0068] S260, obtaining a target image of a target grounding switch, inputting the target image into a grounding switch state determination model, and determining a target opening and closing state of the target grounding switch according to an output result of the grounding switch state determination model.

[0069] S270. Send the target opening and closing state to the automation system, so that the automation system can control the state of the target grounding switch according to the target opening and closing state.

[0070] Among them, the automation system is used to automatically control the opening and closing states of the grounding switch, which can avoid personal risks to operators caused by accidents during the process.

[0071] In an embodiment of the present invention, after the target image is input into the grounding switch state determination model and the target opening and closing state of the target grounding switch is determined according to the target output result, the target opening and closing state can be sent to the automation system to provide the automation system with a reliable basis for judging the opening and closing state of the grounding switch, so that the automation system can accurately control the state of the target grounding switch.

[0072] The technical solution of the embodiment of the present invention comprises the following steps: obtaining a sample set of grounding switch images; inputting the sample set of grounding switch images into a generator of an adversarial network model, and generating a fuzzy image of each grounding switch image in the sample set of grounding switch images after the generator learns the sample set of grounding switch images; inputting the fuzzy image into a discriminator of the adversarial network model, guiding the output of the discriminator through the separation and combination states annotated in the sample set of grounding switch images, iteratively optimizing the network parameters of the adversarial network model, and generating a grounding switch state determination model; obtaining a test set of grounding switch images, and inputting the test set of grounding switch images into the grounding switch state determination model; evaluating the accuracy of the grounding switch model according to the test results output by the grounding switch state determination model; obtaining a target image of a target grounding switch, inputting the target image into the grounding switch state determination model, and determining a target separation and combination state of the target grounding switch according to the output result of the grounding switch state determination model; and sending the target separation and combination state to an automation system, so that the automation system can control the state of the target grounding switch according to the target separation and combination state. The technical solution of the embodiment of the present invention determines the on / off state of the target grounding switch by using the grounding switch state determination model obtained by training the adversarial network model and the acquired target image, thereby improving the efficiency and accuracy of judging the on / off state of the target grounding switch and avoiding the potential safety hazards caused by human eye observation. At the same time, it provides a reliable basis for the automation system that controls the state of the grounding switch.

[0073] Embodiment 3

[0074] Figure 3 This is a schematic diagram of the structure of a device for determining the state of a grounding switch provided in the third embodiment of the present invention. Figure 3 As shown, the device comprises:

[0075] A grounding switch image sample set acquisition module 310 is used to acquire a grounding switch image sample set;

[0076] A state determination model generation module 320 is used to train the adversarial network model according to the grounding switch image sample set to generate a grounding switch state determination model;

[0077] The grounding switch state determination module 330 is used to obtain a target image of a target grounding switch, input the target image into the grounding switch state determination model, and determine the target opening and closing state of the target grounding switch according to the output result of the grounding switch state determination model.

[0078] Optionally, the grounding switch image sample set acquisition module 310 includes:

[0079] A grounding switch image acquisition unit, used to acquire at least two grounding switch images from a grounding switch image library of the substation, and respectively determine the opening and closing states of the grounding switches corresponding to the at least two grounding switch images;

[0080] The grounding switch image sample set generating unit is used to mark the corresponding at least two grounding switch images based on the opening and closing states to generate a grounding switch image sample set.

[0081] Optionally, the state determination model generation module 320 includes:

[0082] A fuzzy image generating unit, used for inputting the grounding switch image sample set into a generator of an adversarial network model, wherein the generator generates a fuzzy image of each grounding switch image in the grounding switch image sample set after learning the grounding switch image sample set;

[0083] The state determination model generation unit is used to input the fuzzy image into the discriminator of the adversarial network model, guide the discriminator output through the open and close states annotated in the grounding switch image sample set, iteratively optimize the network parameters of the adversarial network model, and generate a grounding switch state determination model.

[0084] Optionally, the state determination model generation unit is specifically used to: determine the output error of the adversarial network model according to the target loss function during the iterative optimization of the network parameters of the adversarial network model, until the output error is less than a preset threshold, and generate a grounding switch state determination model; wherein the target loss function includes an adversarial loss function, a conditional loss function, a perceptual loss function, and a gradient loss function.

[0085] Optionally, the state determination model generation module 320 further includes:

[0086] A grounding switch image test set acquisition unit, used for acquiring a grounding switch image test set, and inputting the grounding switch image test set into the grounding switch state determination model;

[0087] The model accuracy evaluation unit is used to evaluate the accuracy of the grounding switch state determination model according to the test result output by the grounding switch state determination model.

[0088] Optionally, the device further comprises:

[0089] The target grounding switch state control module is used to send the target opening and closing state to the automation system, so that the automation system can control the state of the target grounding switch according to the target opening and closing state.

[0090] The grounding switch state determination device provided in the embodiment of the present invention can execute the grounding switch state determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0091] Embodiment 4

[0092] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0093] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0094] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0095] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for determining the state of a grounding knife switch.

[0096] In some embodiments, the grounding switch state determination method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the grounding switch state determination method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the grounding switch state determination method in any other appropriate manner (for example, by means of firmware).

[0097] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0098] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0099] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0100] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0101] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0102] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0103] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0104] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining the state of a grounding switch, characterized in that: The method comprises: Acquire a grounding switch image sample set; The adversarial network model is trained according to the grounding switch image sample set to generate a grounding switch state determination model; A target image of a target grounding switch is acquired, the target image is input into the grounding switch state determination model, and a target opening and closing state of the target grounding switch is determined according to an output result of the grounding switch state determination model.

2. The method according to claim 1, characterized in that Get the grounding switch image sample set, including: Acquire at least two grounding switch images from a grounding switch image library of the substation, and respectively determine the opening and closing states of the grounding switches corresponding to the at least two grounding switch images; Based on the open / closed state, the corresponding at least two grounding switch images are respectively annotated to generate a grounding switch image sample set.

3. The method according to claim 1, characterized in that The adversarial network model is trained according to the grounding switch image sample set to generate a grounding switch state determination model, including: The grounding switch image sample set is input into the generator of the adversarial network model, and after the generator learns the grounding switch image sample set, a fuzzy image of each grounding switch image in the grounding switch image sample set is generated; The fuzzy image is input into the discriminator of the adversarial network model, the discriminator output is guided by the open and close states annotated in the grounding switch image sample set, the network parameters of the adversarial network model are iteratively optimized, and the grounding switch state determination model is generated.

4. The method according to claim 3, characterized in that Iteratively optimizing the network parameters of the adversarial network model to generate a grounding switch state determination model includes: In the process of iteratively optimizing the network parameters of the adversarial network model, the output error of the adversarial network model is determined according to the target loss function until the output error is less than a preset threshold, and a grounding switch state determination model is generated; wherein the target loss function includes an adversarial loss function, a conditional loss function, a perceptual loss function, and a gradient loss function.

5. The method according to claim 3, characterized in that: After iteratively optimizing the network parameters of the adversarial network model to generate a grounding switch state determination model, the method further includes: Acquire a grounding switch image test set, and input the grounding switch image test set into the grounding switch state determination model; According to the test result output by the grounding switch state determination model, the accuracy of the grounding switch state determination model is evaluated.

6. The method according to claim 1, characterized in that After determining the target opening and closing state of the target grounding switch according to the output result of the grounding switch state determination model, the method further includes: The target opening and closing state is sent to an automation system, so that the automation system can control the state of the target grounding switch according to the target opening and closing state.

7. A device for determining the state of a grounding switch, characterized in that: The device comprises: A grounding switch image sample set acquisition module is used to acquire a grounding switch image sample set; A state determination model generation module is used to train the adversarial network model according to the grounding switch image sample set to generate a grounding switch state determination model; The grounding switch state determination module is used to obtain a target image of a target grounding switch, input the target image into the grounding switch state determination model, and determine the target opening and closing state of the target grounding switch according to the output result of the grounding switch state determination model.

8. The device according to claim 7, characterized in that The state determination model generation module comprises: A fuzzy image generating unit, used for inputting the grounding switch image sample set into a generator of an adversarial network model, wherein the generator generates a fuzzy image of each grounding switch image in the grounding switch image sample set after learning the grounding switch image sample set; The state determination model generation unit is used to input the fuzzy image into the discriminator of the adversarial network model, guide the discriminator output through the open and close states annotated in the grounding switch image sample set, iteratively optimize the network parameters of the adversarial network model, and generate a grounding switch state determination model.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the state of a grounding switch according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the state of a grounding switch according to any one of claims 1 to 6 when executing the computer instructions.