Grounding grid corrosion area positioning method, device, equipment, storage medium and program product

Abstract

The invention discloses a grounding grid corrosion area positioning method, device and equipment, a storage medium and a program product, and the method comprises the steps: determining an initial electrical impedance imaging image corresponding to a grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage and a resistance value of a sampling resistor; inputting the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network by using the historical initial electrical impedance imaging image; and inputting the reconstructed electrical impedance imaging image into the target image segmentation model to obtain a corrosion area image corresponding to the grounding grid. By adopting the method, the accuracy of positioning the corrosion area of the grounding grid can be improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a method, device, equipment, storage medium and program product for locating a corrosion area of ​​a grounding grid. Background Art

[0002] The grounding grid is an important infrastructure to ensure the safe and reliable operation of the power system. As it is buried underground all year round, it is affected by many factors such as soil corrosion and microorganisms, causing it to corrode or even break, threatening the safety of power transmission and transformation workers and electrical equipment.

[0003] Electrical Impedance Tomography (EIT) is an imaging method that reconstructs the internal conductivity distribution of an object by measuring the potential difference on the surface of the object. Applying EIT technology to the field of grounding grid corrosion diagnosis can reconstruct the grounding grid on-site image and display the grounding grid corrosion location without large-scale excavation. However, due to the low imaging quality and low image resolution, some minor corrosion points cannot be effectively identified, resulting in inaccurate positioning of the grounding grid corrosion area.

[0004] Therefore, how to improve the accuracy of locating the corrosion area of ​​the grounding grid has become an urgent problem to be solved. Summary of the invention

[0005] The embodiments of the present application provide a method, device, equipment, storage medium and program product for locating a corrosion area of ​​a grounding grid, which can improve the accuracy of locating a corrosion area of ​​a grounding grid.

[0006] In a first aspect, an embodiment of the present application provides a method for locating a grounding grid corrosion area, the method comprising:

[0007] Determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor;

[0008] The initial electrical impedance imaging image is input into the target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0009] The reconstructed electrical impedance tomography image is input into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0010] In one embodiment, the target image segmentation model includes a feature extraction network, a region proposal network and an image segmentation network; the reconstructed electrical impedance imaging image is input into the target image segmentation model to obtain a corrosion area image corresponding to the grounding grid, including: inputting the reconstructed electrical impedance imaging image into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image; inputting the feature map into the region proposal network to obtain candidate area coordinates; inputting the feature map and the candidate area coordinates into the image segmentation network to obtain a corrosion area image corresponding to the grounding grid.

[0011] In one embodiment, the feature extraction network includes a residual subnetwork, a convolutional attention subnetwork and a feature pyramid subnetwork; the reconstructed electrical impedance imaging image is input into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image, including: inputting the reconstructed electrical impedance imaging image into the residual subnetwork to obtain a multi-scale feature map corresponding to the reconstructed electrical impedance imaging image; inputting the multi-scale feature map into the convolutional attention subnetwork to obtain a local feature map corresponding to the reconstructed electrical impedance imaging image; inputting the local feature map into the feature pyramid subnetwork to obtain a feature map corresponding to the reconstructed electrical impedance imaging image.

[0012] In one embodiment, the target generative adversarial network is obtained in the following manner: based on the historical target detection voltage, historical reference voltage and historical sampling resistance of the grounding grid obtained by historical acquisition, a historical initial electrical impedance imaging image is generated; the historical initial electrical impedance imaging image is input into the generator of the generative adversarial network to generate an adversarial sample image corresponding to the historical initial electrical impedance imaging image, and the adversarial sample image is input into the discriminator of the generative adversarial network to obtain a predicted category of the adversarial sample image; based on the predicted category of the adversarial sample image and the actual category of the historical initial electrical impedance imaging image, an evaluation index value corresponding to the generative adversarial network is determined; based on the evaluation index value determined in the i-th iteration and the generated adversarial sample image, the generative adversarial network of the i+1th iteration is iteratively updated until a preset condition for stopping the generation of the adversarial sample image is met, and the generative adversarial network of the last iteration is used as the target generative adversarial network; wherein i is a positive integer greater than or equal to 1.

[0013] In one of the embodiments, an initial electrical impedance imaging image corresponding to the ground grid is determined based on a reference voltage corresponding to the ground grid, a target detection voltage, and a resistance value of a sampling resistor, including: determining an excitation current corresponding to the ground grid based on the reference voltage corresponding to the ground grid and the resistance value of the sampling resistor; determining the resistance of the surface of the ground grid based on the target detection voltage and the excitation current corresponding to the ground grid; determining the conductivity of the surface of the ground grid based on the resistance of the surface of the ground grid; and determining an initial electrical impedance imaging image corresponding to the ground grid based on the conductivity of the surface of the ground grid.

[0014] In one of the embodiments, the target detection voltage is determined in the following manner: obtaining an initial voltage signal of the collected grounding grid; extracting a voltage signal within a preset frequency range from the initial voltage signal as an impedance imaging voltage signal; obtaining a target detection voltage signal based on the impedance imaging voltage signal; and converting the target detection voltage signal using an analog-to-digital converter to obtain a target detection voltage.

[0015] In a second aspect, the present application provides a grounding grid corrosion area positioning device, the device comprising:

[0016] A determination module, used to determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor;

[0017] A processing module is used to input the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0018] The processing module is also used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0019] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0020] A determination module, used to determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor;

[0021] A processing module is used to input the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0022] The processing module is also used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0023] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0024] A determination module, used to determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor;

[0025] A processing module is used to input the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0026] The processing module is also used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0027] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0028] A determination module, used to determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor;

[0029] A processing module is used to input the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0030] The processing module is also used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0031] The above-mentioned method, device, equipment, storage medium and program product for locating the corrosion area of ​​the grounding grid, the computer equipment can determine the initial impedance imaging image corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid, the target detection voltage and the resistance value of the sampling resistor; the initial impedance imaging image is input into the target generative adversarial network to obtain the reconstructed impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial impedance imaging image; the reconstructed impedance imaging image is input into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid. Using this method, the computer equipment can obtain a reconstructed impedance imaging image that is closer to the real image corresponding to the grounding grid by inputting the initial impedance imaging image corresponding to the grounding grid into the pre-trained target generative adversarial network, which can improve the accuracy of the impedance imaging image, thereby, the impedance imaging image that is closer to the real image corresponding to the grounding grid is input into the target image segmentation model, and a more accurate corrosion area image corresponding to the grounding grid can be obtained, and then, the accuracy of contact network corrosion area positioning can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 This is a schematic diagram of an application scenario of a grounding grid corrosion area positioning method provided in an embodiment of the present application;

[0034] Figure 2 It is a schematic diagram of a flow chart of a method for locating a corrosion area of ​​a grounding grid provided in an embodiment of the present application;

[0035] Figure 3 This is a schematic diagram of a data collection process provided by an embodiment of the present application;

[0036] Figure 4 It is a structural schematic diagram of a target image segmentation model provided in an embodiment of the present application;

[0037] Figure 5 It is a flow chart of another method for locating a corrosion area of ​​a grounding grid provided in an embodiment of the present application;

[0038] Figure 6 is a schematic diagram of a feature extraction network provided in an embodiment of the present application;

[0039] Figure 7 Schematic diagram of a network structure of a residual sub-network provided in an embodiment of the present application;

[0040] Figure 8 It is a structural schematic diagram of a contact network corrosion area positioning device provided in an embodiment of the present application;

[0041] Fig. 9 It is a structural schematic diagram of another grounding grid corrosion area positioning device provided in an embodiment of the present application;

[0042] Fig.10 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0044] The following introduces the application scenarios of the grounding grid corrosion area positioning method provided in the embodiments of the present application.

[0045] See also Figure 1 , Figure 1 Schematic diagram of an application scenario of a method for locating a grounding grid corrosion area provided in an embodiment of the present application. Figure 1 As shown, the computer device 101 ( Figure 1 In the figure, the computer device 101 is drawn as an example of a terminal device) and a grounding grid 102.

[0046] Among them, the computer device 101 can first determine the initial electrical impedance imaging image corresponding to the grounding grid 102, and then input the initial electrical impedance imaging image corresponding to the grounding grid 102 into a pre-trained target generation adversarial network to obtain a reconstructed electrical impedance imaging image, and finally, input the reconstructed electrical impedance imaging image into a target image segmentation model to obtain a corrosion area image corresponding to the grounding grid. In this way, the computer device 101 can obtain a reconstructed electrical impedance imaging image by inputting the initial electrical impedance imaging image corresponding to the grounding grid 102 into a pre-trained target generation adversarial network. The reconstructed electrical impedance imaging image is an electrical impedance imaging image that is closer to the real image corresponding to the grounding grid, which can improve the accuracy of the electrical impedance imaging image. Therefore, the electrical impedance imaging image that is closer to the real image corresponding to the grounding grid is input into the target image segmentation model, and a more accurate corrosion area image corresponding to the grounding grid can be obtained, thereby improving the accuracy of positioning the corrosion area of ​​the contact network.

[0047] Optionally, the computer device 101 may be a terminal device or a server. The terminal device mentioned here may include but is not limited to: a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart watch, a smart TV, a smart car terminal, etc. The server mentioned here may be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers.

[0048] See also Figure 2 , Figure 2 1 is a flow chart of a method for locating a grounding grid corrosion area provided in an embodiment of the present application. The method can be executed by a computer device (for example, the above-mentioned computer device 101). Figure 2 As shown, the method for locating the grounding grid corrosion area may include but is not limited to the following steps:

[0049] S201 : Determine an initial electrical impedance imaging image corresponding to the ground grid based on a reference voltage corresponding to the ground grid, a target detection voltage, and a resistance value of a sampling resistor.

[0050] The reference voltage corresponding to the ground grid refers to the reference voltage obtained by the excitation electrodes for determining the initial electrical impedance tomography image. The excitation electrodes include multiple groups (eg, 16 groups) of electrodes, and the current of each group of electrodes is a known value.

[0051] Among them, the target detection voltage corresponding to the grounding grid refers to the detection voltage used to determine the initial electrical impedance imaging image, and the voltage signal corresponding to the target detection voltage refers to the initial voltage signal corresponding to the frequency band below 100KHz. The initial voltage signal is obtained by collecting the voltage on the surface of the grounding grid using measuring electrodes.

[0052] Combine the following Figure 3 For an introduction to the reference voltage corresponding to the ground grid, the target detection voltage, and the determination method of the sampling resistor value, see Figure 3 , Figure 3 It is a schematic diagram of a data collection process provided in an embodiment of the present application.

[0053] Among them, Field-Programmable Gate Array (FPGA) is a product further developed on the basis of programmable array logic (PAL), generic array logic (GAL) and other programmable devices. It appears as a semi-custom circuit in the field of Application Specific Integrated Circuit (ASIC), which not only solves the shortcomings of custom circuits, but also overcomes the shortcomings of the limited number of gate circuits of the original programmable devices.

[0054] Direct Digital Synthesizer (DDS) is a new type of frequency synthesis technology with advantages such as large bandwidth, short frequency conversion time, high resolution and good phase continuity. It is easy to realize digital control modulation of frequency, phase and amplitude, and is widely used in the communication field.

[0055] An analog-to-digital converter (ADC) is a device used to convert a continuous signal in analog form into a discrete signal in digital form. An analog-to-digital converter can provide a signal for measurement.

[0056] like Figure 3As shown, the computer device can first control the DDS through FPGA to generate a sinusoidal voltage signal; secondly, since the signal frequency required for electrical impedance imaging is generally 10KHz-100KHz, a low-pass filter is used to extract the voltage signal corresponding to the frequency band below 100KHz from the sinusoidal voltage signal; then, since electrical impedance imaging needs to form an electromagnetic field in the grounding grid by current excitation, the voltage signal corresponding to the frequency band below 100KHz is converted into a current signal through a voltage-controlled constant current source; then, the grounding grid is subjected to current excitation under the action of an analog switch, and a reference voltage signal corresponding to the grounding grid is obtained through a sampling resistor in the current loop; finally, ADC1 is used to convert the reference voltage signal corresponding to the grounding grid into a reference voltage corresponding to the grounding grid.

[0057] The computer equipment can also obtain the initial voltage signal on the surface of the grounding grid under the action of the electrodes, and select different groups of electrodes for differential amplification through analog switches. Before differential amplification, the 10KHz-100KHz electrical impedance imaging voltage signal is extracted from the initial voltage signal through a bandpass filter suitable for electrical impedance imaging. After that, the different groups of electrodes are program-controlled amplified using a resistor network plus an operational amplifier to obtain the target detection voltage signal corresponding to the grounding grid. Finally, ADC2 is used to convert the target detection voltage signal corresponding to the grounding grid into the target detection voltage.

[0058] S202: Input the initial electrical impedance imaging image into the target generative adversarial network to obtain a reconstructed electrical impedance imaging image.

[0059] Among them, the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance tomography images.

[0060] Among them, the generative adversarial network is a deep learning model composed of a generator and a discriminator, which compete with each other to learn and generate new samples similar to training data. Among them, the generator is used to accept random noise or latent space vectors as input and map them to the data space to generate fake samples. The others are usually composed of a series of deconvolution (or transposed convolution) layers and activation functions, which are used to map the latent space vectors to the data space and generate realistic samples; the discriminator is used to accept real samples (from training data) and fake samples generated by the generator, and try to distinguish them. The others are usually composed of a series of convolution layers and activation functions, which are used to extract features from the input samples and output a probability, indicating the probability that the input sample is a real sample.

[0061] It should be noted that the computer device inputs the initial electrical impedance imaging image into the target generative adversarial network to obtain the reconstructed electrical impedance imaging image, which means that the initial electrical impedance imaging image is input into the generator included in the target generative adversarial network to obtain the reconstructed electrical impedance imaging image.

[0062] S203, inputting the reconstructed electrical impedance tomography image into the target image segmentation model to obtain a corrosion area image corresponding to the grounding grid.

[0063] In an optional implementation, the target image segmentation model may be a machine learning algorithm model or a deep learning algorithm model, which is not limited here. For example, the target image segmentation model is a Mask R-CNN algorithm (a target detection / segmentation algorithm that can effectively detect targets in an image) with an added convolutional attention module.

[0064] In an embodiment of the present application, the computer device can determine the initial impedance imaging image corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid, the target detection voltage and the resistance value of the sampling resistor; the initial impedance imaging image is input into the target generative adversarial network to obtain a reconstructed impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial impedance imaging image; the reconstructed impedance imaging image is input into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid. Using this method, the computer device can obtain a reconstructed impedance imaging image that is closer to the real image corresponding to the grounding grid by inputting the initial impedance imaging image corresponding to the grounding grid into the pre-trained target generative adversarial network, which can improve the accuracy of the impedance imaging image, thereby, the impedance imaging image that is closer to the real image corresponding to the grounding grid is input into the target image segmentation model, and a more accurate corrosion area image corresponding to the grounding grid can be obtained, thereby improving the accuracy of the contact network corrosion area positioning.

[0065] In an optional embodiment, Figure 2 In the ground grid corrosion area positioning method shown in the figure, the target image segmentation model includes a feature extraction network, a region proposal network and an image segmentation network. Figure 4 The target image segmentation model provided in the embodiment of the present application is introduced. Figure 4 It is a structural schematic diagram of a target image segmentation model provided in an embodiment of the present application.

[0066] like Figure 4 As shown, the target image segmentation model may include a feature extraction network 401, a region proposal network 402 and an image segmentation network 403.

[0067] Among them, the feature extraction network 401 includes a residual subnetwork 4011, a convolutional attention subnetwork 4012 and a feature pyramid subnetwork 4013; the residual subnetwork 4011 is used to extract the multi-scale feature map corresponding to the reconstructed electrical impedance imaging image; the convolutional attention subnetwork 4012 is used to obtain the local feature map corresponding to the reconstructed electrical impedance imaging image from the multi-scale feature map; the feature pyramid subnetwork 4013 is used to obtain the feature map corresponding to the reconstructed electrical impedance imaging image based on the local feature map.

[0068] The region proposal network 402 is used to obtain candidate region coordinates based on the feature map corresponding to the reconstructed electrical impedance tomography image.

[0069] The image segmentation network 403 includes a region of interest alignment network 4031, a head network 4032, and a fully connected layer 4033. The image segmentation network 403 is used to determine the corrosion region image corresponding to the grounding grid based on the feature map and candidate region coordinates corresponding to the reconstructed electrical impedance imaging image.

[0070] Combine the following Figure 4 and Figure 5 , a method for locating the corrosion area of ​​a grounding grid provided in this application is described.

[0071] See also Figure 5 , Figure 5 1 is a flow chart of another method for locating a grounding grid corrosion area provided in an embodiment of the present application. Figure 2 The difference between the method of locating the corroded area of ​​the grounding grid shown is that Figure 5 The method shown specifically describes how the computer device determines the initial impedance imaging image corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid, the target detection voltage and the resistance value of the sampling resistor, and how to input the reconstructed impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid. Figure 5 As shown, the method for locating the grounding grid corrosion area may include but is not limited to the following steps:

[0072] S501 : Determine an excitation current corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid and a resistance value of a sampling resistor.

[0073] For example, see Figure 3 The data collection process diagram is shown in Figure 3As shown, the computer device can first control the DDS through FPGA to generate a sinusoidal voltage signal; secondly, since the signal frequency required for electrical impedance imaging is generally 10KHz-100KHz, a low-pass filter is used to extract the voltage signal corresponding to the frequency band below 100KHz from the sinusoidal voltage signal; then, since electrical impedance imaging needs to form an electromagnetic field in the grounding grid by current excitation, the voltage signal corresponding to the frequency band below 100KHz is converted into a current signal through a voltage-controlled constant current source; then, the grounding grid is subjected to current excitation under the action of an analog switch, and a reference voltage signal corresponding to the grounding grid is obtained through a sampling resistor in the current loop; finally, ADC1 is used to convert the reference voltage signal corresponding to the grounding grid into a reference voltage corresponding to the grounding grid.

[0074] The resistance of the sampling resistor is known. The computer device determines the excitation current corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid and the resistance of the sampling resistor. The excitation current corresponding to the grounding grid can be obtained by dividing the reference voltage corresponding to the grounding grid by the resistance of the sampling resistor.

[0075] S502 : Determine the surface resistance of the grounding grid based on the target detection voltage and the excitation current corresponding to the grounding grid.

[0076] In an optional embodiment, the target detection voltage can be determined by a computer device in the following manner: acquiring an initial voltage signal of the collected grounding grid; extracting a voltage signal within a preset frequency range from the initial voltage signal as an impedance imaging voltage signal; obtaining a target detection voltage signal based on the impedance imaging voltage signal; and converting the target detection voltage signal using an analog-to-digital converter to obtain a target detection voltage.

[0077] For example, see Figure 3 The data collection process diagram is shown in Figure 3 As shown, the computer device can obtain the initial voltage signal on the surface of the grounding grid under the action of the electrode, and select different groups of electrodes for differential amplification through analog switches. Before differential amplification, the 10KHz-100KHz electrical impedance imaging voltage signal is extracted from the initial voltage signal through a bandpass filter suitable for electrical impedance imaging. After that, the different groups of electrodes are program-controlled amplified using a resistor network plus an operational amplifier to obtain the target detection voltage signal corresponding to the grounding grid. Finally, ADC2 is used to convert the target detection voltage signal corresponding to the grounding grid into the target detection voltage.

[0078] The computer device determines the surface resistance of the grounding grid based on the target detection voltage and excitation current corresponding to the grounding grid, and can obtain the surface resistance of the grounding grid by dividing the target detection voltage corresponding to the grounding grid by the excitation current.

[0079] S503: Determine the conductivity of the surface of the grounding grid based on the resistance of the surface of the grounding grid.

[0080] The conductivity of the grounding grid surface is the reciprocal of the resistance of the grounding grid surface.

[0081] S504: Determine an initial electrical impedance imaging image corresponding to the grounding grid based on the electrical conductivity of the grounding grid surface.

[0082] S505: Input the initial electrical impedance imaging image into the target generative adversarial network to obtain a reconstructed electrical impedance imaging image.

[0083] Among them, the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance tomography images.

[0084] In the present application, the computer device inputs the initial impedance imaging image corresponding to the grounding grid into the target generative adversarial network, which can generate an impedance image that is closer to the actual image corresponding to the grounding grid. In other words, the accuracy of the impedance image can be improved, which is conducive to determining a more accurate corrosion area image corresponding to the grounding grid based on a more accurate impedance image, and further, it is conducive to improving the accuracy of positioning the corrosion area of ​​the grounding grid.

[0085] S506: Input the reconstructed electrical impedance imaging image into a feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image.

[0086] In an optional embodiment, the feature extraction network includes a residual subnetwork, a convolutional attention subnetwork and a feature pyramid subnetwork; the computer device inputs the reconstructed electrical impedance imaging image into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image, which may include: inputting the reconstructed electrical impedance imaging image into the residual subnetwork to obtain a multi-scale feature map corresponding to the reconstructed electrical impedance imaging image; inputting the multi-scale feature map into the convolutional attention subnetwork to obtain a local feature map corresponding to the reconstructed electrical impedance imaging image; inputting the local feature map into the feature pyramid subnetwork to obtain a feature map corresponding to the reconstructed electrical impedance imaging image.

[0087] For example, see Figure 6 , Figure 6 Schematic diagram of a feature extraction network provided in an embodiment of the present application. Figure 6 As shown in the figure, it includes the residual sub-network (also called Res2Net), the convolutional attention sub-network (ConvolutionalBlock Attention Module, CBAM) and the feature pyramid sub-network.

[0088] like Figure 6As shown, the computer device first uses the convolutional layers (Conv1~Conv5) in the residual sub-network to extract the feature maps of the input reconstructed electrical impedance imaging image from bottom to top, and represents the output feature maps as C1, C2, C3, C4, and C5; secondly, a 1×1 convolution fusion is performed to change the dimension of the feature map, and the convolutional attention mechanism sub-network CBAM is used to automatically identify the corrosion area to improve the feature extraction ability of the model; then, the feature pyramid network is used to perform convolution fusion with the side features through top-down upsampling, and a 3×3 convolution is used to convolve each fusion result to eliminate the aliasing effect of upsampling, and finally the convolution results are fused to obtain P2~P5, and combined with the top-level feature map P6 to obtain the feature map corresponding to the reconstructed electrical impedance image, thereby realizing multi-scale fusion of features.

[0089] For example, see Figure 7 , Figure 7 Schematic diagram of a residual sub-network provided in an embodiment of the present application. Figure 7 As shown in the figure, the residual subnetwork can also be called Res2Net, which represents multi-scale features at a finer granularity and increases the receptive fields of each network layer.

[0090] S507: Input the feature map into the region proposal network to obtain the coordinates of the candidate region.

[0091] Among them, the Region Proposal Network (RPN) is a key component in deep learning object detection, which uses feature maps to generate candidate boxes. The RPN input is the preprocessed feature map, and the output includes the object existence probability and the candidate box coordinates.

[0092] S508: Input the feature map and the candidate region coordinates into the image segmentation network to obtain the corrosion region image corresponding to the grounding grid.

[0093] Among them, the image segmentation network is used to determine the corrosion area image corresponding to the grounding grid based on the feature map and candidate area coordinates corresponding to the reconstructed electrical impedance imaging image.

[0094] In an optional implementation, the image segmentation network includes a region of interest alignment network, a head network, and a fully connected layer.

[0095] In an embodiment of the present application, the computer device can obtain a multi-scale feature map corresponding to the reconstructed impedance imaging image by inputting the reconstructed impedance imaging image into a feature extraction network; then, the multi-scale feature map is input into a region proposal network to improve the accuracy of the determined candidate region coordinates; finally, the multi-scale feature map and the candidate region coordinates are input into an image segmentation network to improve the accuracy of the contact area image corresponding to the determined grounding grid, thereby improving the accuracy of the positioning of the corrosion area of ​​the grounding grid.

[0096] In an optional embodiment, Figure 2 and Figure 5 In the ground grid corrosion area positioning method shown, the target generation adversarial network can be obtained by a computer device in the following manner:

[0097] Step 1: Generate a historical initial electrical impedance imaging image based on the historical target detection voltage, historical reference voltage and historical sampling resistor value of the grounding grid obtained through historical collection.

[0098] Step 2: Input the historical initial electrical impedance imaging image into the generator of the generative adversarial network to generate an adversarial sample image corresponding to the historical initial electrical impedance imaging image, and input the adversarial sample image into the discriminator of the generative adversarial network to obtain the predicted category of the adversarial sample image.

[0099] Step 3: Based on the predicted category of the adversarial sample image and the actual category of the historical initial electrical impedance tomography image, determine the evaluation index value corresponding to the generative adversarial network.

[0100] In an optional implementation, the evaluation index values ​​include but are not limited to accuracy, precision, loss value, etc.

[0101] Step 4: Based on the evaluation index value determined in the i-th iteration and the generated adversarial sample image, iteratively update the generative adversarial network of the i+1-th iteration until the preset condition for stopping generating adversarial sample images is met, and use the generative adversarial network of the last iteration as the target generative adversarial network; wherein i is a positive integer greater than or equal to 1.

[0102] Among them, the preset condition for stopping generating adversarial sample images is related to the evaluation index value. Exemplarily, assuming that the evaluation index value is accuracy, the preset condition for stopping generating adversarial sample images may be that the accuracy corresponding to the generated adversarial network is greater than or equal to the preset accuracy threshold. Assuming that the evaluation index value is precision, the preset condition for stopping generating adversarial sample images may be that the precision corresponding to the generated adversarial network is greater than or equal to the preset precision threshold. Assuming that the evaluation index value is loss value, the preset condition for stopping generating adversarial sample images may be that the loss value corresponding to the generated adversarial network is less than or equal to the preset loss threshold.

[0103] By adopting this implementation mode, the computer device iteratively updates the initialized generative adversarial network by utilizing the historical initial electrical impedance imaging image corresponding to the grounding grid and the actual category of the historical initial electrical impedance imaging image, and uses the generative adversarial network of the last iteration as the target generative adversarial network when the preset condition for stopping generating adversarial sample images is met. This is beneficial for the computer device to input the initial electrical impedance imaging image corresponding to the grounding grid into the target generative adversarial network during actual application to generate an electrical impedance image that is closer to the actual image corresponding to the grounding grid. That is to say, the accuracy of the electrical impedance image can be improved, thereby facilitating the determination of a more accurate corrosion area image corresponding to the grounding grid based on a more accurate electrical impedance image, and further, facilitating the improvement of the accuracy of positioning the corrosion area of ​​the grounding grid.

[0104] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0105] Based on the same inventive concept, the embodiment of the present application also provides a grounding grid corrosion area positioning device for implementing the grounding grid corrosion area positioning method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more grounding grid corrosion area positioning device embodiments provided below can refer to the limitations of the grounding grid corrosion area positioning method above, and will not be repeated here.

[0106] See also Figure 8 , Figure 8 Schematic diagram of the structure of a contact network corrosion area positioning device provided in an embodiment of the present application. Figure 8 As shown, the device includes an excitation electrode 801, a measuring electrode 802, an impedance imaging data acquisition module 803, a wireless transmission module 804, an impedance imaging image reconstruction module 805, a database module 806, a grounding grid corrosion positioning module 807, and a cloud platform 808.

[0107] Among them, the computer equipment can use the impedance imaging data acquisition module 803 to extract the target detection voltage and reference voltage of the grounding grid area to be tested by using the 16-electrode impedance imaging image acquisition method based on the excitation electrode 801 and the measuring electrode 802; then, the conductivity image data (i.e., the initial impedance imaging image) is obtained by calculation; then, the image data is transmitted to the impedance imaging image reconstruction module 805 by using the wireless transmission module 804, and the image is edge enhanced and denoised by using the generative adversarial network model to obtain a grounding grid impedance imaging image with higher imaging spatial resolution and fewer edge blur artifacts (i.e., the reconstructed impedance imaging image); then, the grounding grid corrosion positioning module 807 is used to perform image segmentation on the reconstructed impedance imaging image to realize the automatic recognition and positioning function.

[0108] The detailed description of the electrical impedance imaging data acquisition module 803 can be found in the above description. Figure 3 The description will not be repeated here.

[0109] Among them, the wireless transmission module 804 can be composed of a long-range (Lora) module, which transmits the data of the electrical impedance imaging image reconstruction module 805 to the terminal router, and then uploads it to the database 806 and the cloud platform 808 through the wireless local area network (Wireless Fidelity, WIFI). Among them, the Lora module is a radio frequency module capable of long-distance low-power communication, and uses Lora technology to achieve wireless communication. Lora technology is a physical layer wireless digital communication modulation technology, also known as spread spectrum continuous frequency modulation technology (Chirp Modulation). Its characteristic is that it can propagate farther than other wireless methods at the same power consumption, achieving the unity of low power consumption and long distance, and expanding the communication distance of traditional wireless communication radio frequency by 3-5 times at the same power consumption.

[0110] The electrical impedance imaging image reconstruction module 805 is used to achieve image reconstruction using a generative adversarial network to obtain a high-quality electrical impedance imaging image of the grounding grid.

[0111] The grounding grid corrosion positioning module 807 is used to use the Mask R-CNN algorithm with an added convolutional attention module to perform image segmentation processing on the impedance image output by the impedance imaging image reconstruction module 805, so as to accurately identify the corrosion area image corresponding to the contact network.

[0112] See also Fig. 9 , Fig. 9 Schematic diagram of another grounding grid corrosion area positioning device provided in an embodiment of the present application. Fig. 9 As shown, the grounding grid corrosion area positioning device may include but is not limited to:

[0113] A determination module 901 is used to determine an initial electrical impedance imaging image corresponding to the ground grid based on a reference voltage corresponding to the ground grid, a target detection voltage and a resistance value of a sampling resistor;

[0114] Processing module 902, used for inputting the initial electrical impedance imaging image into the target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image;

[0115] The processing module 902 is further used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid.

[0116] In one embodiment, the target image segmentation model includes a feature extraction network, a region proposal network and an image segmentation network; when the processing module 902 is used to input the reconstructed electrical impedance imaging image into the target image segmentation model to obtain the corrosion area image corresponding to the ground grid, it is specifically used to: input the reconstructed electrical impedance imaging image into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image; input the feature map into the region proposal network to obtain candidate area coordinates; input the feature map and candidate area coordinates into the image segmentation network to obtain the corrosion area image corresponding to the ground grid.

[0117] In one embodiment, the feature extraction network includes a residual subnetwork, a convolutional attention subnetwork and a feature pyramid subnetwork; when the processing module 902 is used to input the reconstructed electrical impedance imaging image into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image, it is specifically used to: input the reconstructed electrical impedance imaging image into the residual subnetwork to obtain a multi-scale feature map corresponding to the reconstructed electrical impedance imaging image; input the multi-scale feature map into the convolutional attention subnetwork to obtain a local feature map corresponding to the reconstructed electrical impedance imaging image; input the local feature map into the feature pyramid subnetwork to obtain a feature map corresponding to the reconstructed electrical impedance imaging image.

[0118] In one embodiment, the device may also include a training module, which is used to: generate a historical initial electrical impedance imaging image based on the historical target detection voltage, historical reference voltage and historical sampling resistor value of the grounding grid obtained through historical acquisition; input the historical initial electrical impedance imaging image into the generator of the generative adversarial network to generate an adversarial sample image corresponding to the historical initial electrical impedance imaging image, and input the adversarial sample image into the discriminator of the generative adversarial network to obtain a predicted category of the adversarial sample image; determine an evaluation index value corresponding to the generative adversarial network based on the predicted category of the adversarial sample image and the actual category of the historical initial electrical impedance imaging image; iteratively update the generative adversarial network of the i+1th iteration based on the evaluation index value determined in the i-th iteration and the generated adversarial sample image until the preset condition for stopping the generation of the adversarial sample image is met, and use the generative adversarial network of the last iteration as the target generative adversarial network; wherein i is a positive integer greater than or equal to 1.

[0119] In one embodiment, when the determination module 901 is used to determine the initial electrical impedance imaging image corresponding to the ground grid based on the reference voltage corresponding to the ground grid, the target detection voltage and the resistance value of the sampling resistor, it is specifically used to: determine the excitation current corresponding to the ground grid based on the reference voltage corresponding to the ground grid and the resistance value of the sampling resistor; determine the resistance of the surface of the ground grid based on the target detection voltage and the excitation current corresponding to the ground grid; determine the conductivity of the surface of the ground grid based on the resistance of the surface of the ground grid; determine the initial electrical impedance imaging image corresponding to the ground grid based on the conductivity of the surface of the ground grid.

[0120] In one embodiment, the determination module 903 is also used to: obtain the collected initial voltage signal of the grounding grid; extract the voltage signal within a preset frequency range from the initial voltage signal as an impedance imaging voltage signal; obtain a target detection voltage signal based on the impedance imaging voltage signal; and convert the target detection voltage signal using an analog-to-digital converter to obtain a target detection voltage.

[0121] Each module in the above-mentioned grounding grid corrosion area positioning device can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the terminal device in the form of hardware, or can be stored in the memory of the terminal device in the form of software, so that the processor can call and execute the corresponding operations of the above modules.

[0122] In an exemplary embodiment, the present application provides a computer device, which may be a terminal device, and its internal structure diagram may be as follows: Fig.10As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for locating a grounding grid corrosion area is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0123] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] In an exemplary embodiment, the present application provides a computer device including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned grounding grid corrosion area locating method when executing the computer program.

[0125] In an exemplary embodiment, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the above-mentioned method for locating the corrosion area of ​​the grounding grid are implemented.

[0126] In an exemplary embodiment, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps in the above-mentioned grounding grid corrosion area locating method.

[0127] It should be noted that the data involved in this application (including but not limited to the reference voltage corresponding to the grounding grid, the target detection voltage and the resistance value of the sampling resistor, the initial impedance imaging image, the reconstructed impedance imaging image, the corrosion area image, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0128] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0129] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0130] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for locating a grounding grid corrosion area, characterized in that: The method comprises: Determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor; Inputting the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image; The reconstructed electrical impedance tomography image is input into a target image segmentation model to obtain a corrosion area image corresponding to the grounding grid.

2. The method according to claim 1, characterized in that The target image segmentation model includes a feature extraction network, a region proposal network and an image segmentation network; the reconstructed electrical impedance imaging image is input into the target image segmentation model to obtain the corrosion area image corresponding to the grounding grid, including: Inputting the reconstructed electrical impedance imaging image into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image; Inputting the feature map into the region proposal network to obtain candidate region coordinates; The feature map and the candidate region coordinates are input into the image segmentation network to obtain a corrosion region image corresponding to the grounding grid.

3. The method according to claim 2, characterized in that The feature extraction network includes a residual subnetwork, a convolutional attention subnetwork and a feature pyramid subnetwork; the reconstructed electrical impedance imaging image is input into the feature extraction network to obtain a feature map corresponding to the reconstructed electrical impedance imaging image, including: Inputting the reconstructed electrical impedance tomography image into the residual subnetwork to obtain a multi-scale feature map corresponding to the reconstructed electrical impedance tomography image; Inputting the multi-scale feature map into the convolutional attention sub-network to obtain a local feature map corresponding to the reconstructed electrical impedance imaging image; The local feature map is input into the feature pyramid sub-network to obtain a feature map corresponding to the reconstructed electrical impedance tomography image.

4. The method according to claim 1, characterized in that: The target generative adversarial network is obtained by: Generate a historical initial electrical impedance imaging image based on the historical target detection voltage, historical reference voltage and historical sampling resistor value of the grounding grid obtained through historical collection; Inputting the historical initial electrical impedance tomography image into a generator of a generative adversarial network to generate an adversarial sample image corresponding to the historical initial electrical impedance tomography image, and inputting the adversarial sample image into a discriminator of the generative adversarial network to obtain a predicted category of the adversarial sample image; Determining an evaluation index value corresponding to the generative adversarial network based on the predicted category of the adversarial sample image and the actual category of the historical initial electrical impedance tomography image; Based on the evaluation index value determined in the i-th iteration and the generated adversarial sample image, the generative adversarial network of the i+1-th iteration is iteratively updated until the preset condition for stopping the generation of adversarial sample images is met, and the generative adversarial network of the last iteration is used as the target generative adversarial network; wherein i is a positive integer greater than or equal to 1.

5. The method according to claim 1, characterized in that The determining of the initial electrical impedance imaging image corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid, the target detection voltage and the resistance value of the sampling resistor comprises: Determine the excitation current corresponding to the grounding grid based on the reference voltage corresponding to the grounding grid and the resistance value of the sampling resistor; Determining the resistance of the surface of the grounding grid based on the target detection voltage corresponding to the grounding grid and the excitation current; Determining the conductivity of the surface of the grounding grid based on the resistance of the surface of the grounding grid; Based on the electrical conductivity of the surface of the grounding grid, an initial electrical impedance imaging image corresponding to the grounding grid is determined.

6. The method according to any one of claims 1 to 5, characterized in that: The target detection voltage is determined by: Acquiring the collected initial voltage signal of the grounding grid; Extracting a voltage signal within a preset frequency range from the initial voltage signal as an electrical impedance imaging voltage signal; Based on the electrical impedance imaging voltage signal, obtaining the target detection voltage signal; The target detection voltage signal is converted by using an analog-to-digital converter to obtain the target detection voltage.

7. A grounding grid corrosion area positioning device, characterized in that: The device comprises: A determination module, configured to determine an initial electrical impedance imaging image corresponding to the grounding grid based on a reference voltage corresponding to the grounding grid, a target detection voltage, and a resistance value of a sampling resistor; A processing module, used for inputting the initial electrical impedance imaging image into a target generative adversarial network to obtain a reconstructed electrical impedance imaging image; the target generative adversarial network is obtained by iteratively updating the initial generative adversarial network using the historical initial electrical impedance imaging image; The processing module is also used to input the reconstructed electrical impedance imaging image into a target image segmentation model to obtain a corrosion area image corresponding to the grounding grid.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images