A method for monitoring transformer working scene of power distribution network

Abstract

The present invention relates to a method for monitoring the transformer working scene of a distribution network, comprising: using a distributed identification device to visually determine whether the human head and protective glasses are in a touching state based on various visual information of the human head and various visual information of protective glasses, so as to issue a glasses wearing detection instruction or a glasses wearing non-detection instruction; using a wireless transmission interface to wirelessly transmit the glasses wearing non-detection instruction and the transformer number corresponding to the transformer to a distribution network monitoring node when receiving the glasses wearing non-detection instruction. The method for monitoring the transformer working scene of a distribution network of the present invention uses a distributed identification device with a customized structure to realize a visual judgment of the touching state of the human head and protective glasses at the transformer working site used in the distribution network, that is, whether the human body is wearing protective glasses, thereby improving the management level of the transformer working scene of the distribution network.

Description

Technical Field

[0001] The present invention relates to the field of power distribution networks, and in particular to a method for monitoring a transformer working scene of a power distribution network. Background Art

[0002] A transformer is a device that uses the principle of electromagnetic induction to change AC voltage. Its main components are the primary coil, the secondary coil, and the iron core (magnetic core). In electrical equipment and wireless circuits, it is often used to increase or decrease voltage, match impedance, and provide safety isolation. In a generator, whether the coil moves through a magnetic field or the magnetic field moves through a fixed coil, an electric potential can be induced in the coil. In both cases, the value of the magnetic flux remains unchanged, but the amount of magnetic flux interlinked with the coil changes. This is the principle of mutual induction. A transformer is a device that uses electromagnetic mutual induction to transform voltage, current, and impedance.

[0003] At the transformer work site, maintenance personnel or managers need to wear special protective glasses when entering to avoid damage to the human eyes caused by discharge or other faults of various equipment on site. However, due to the customization of the type of protective glasses and the need for a high-precision analysis mechanism for the relative distribution state of the human body and the protective glasses, there is a lack of detection solutions in the prior art for whether maintenance personnel or managers at the transformer work site are wearing special protective glasses.

[0004] CN117849498A discloses a grid transformer defect detection method and system based on multi-target recognition. Based on the existence status information of all transformers within the grid, all transformers are divided into several transformer clusters, and based on the distribution location information of all transformer clusters within the grid, the grid is divided into several grid sub-areas, so as to realize the zoning detection of all transformers within the grid; it also judges whether the grid sub-area is in the extreme working state, so as to detect the working state of the sensor cluster within the grid sub-area, and ensure that the sensor cluster can cover the detection of the grid sub-area; then, the multi-target detection data generated by the sensor cluster is analyzed, and a working state prediction model for the grid sub-area is constructed, so as to determine the transformer with fault within the grid sub-area, and generate a grid transformer defect location map, so as to realize effective and accurate early warning and troubleshooting of the transformer.

[0005] CN117829437A discloses a fuzzy comprehensive evaluation method for distribution network transformers based on optimized weights and parameters, and the steps are as follows: establish a data set based on multivariate monitoring data of distribution network transformers, select a distribution network transformer state evaluation index set and an evaluation set; establish a transformer comprehensive evaluation membership model and determine the fuzzy comprehensive evaluation weights; construct an optimization objective function based on the deviation value between the actual life and the predicted life, and reversely optimize and adjust the membership model parameters and the fuzzy comprehensive evaluation weights through the data set; determine the fuzzy comprehensive evaluation value of the distribution network transformer to be evaluated based on the optimized parameters and weights.

[0006] CN117540295A discloses a method and device for evaluating the operating status of a distribution network transformer, the method comprising the following steps: acquiring transformer historical data; constructing a feature extraction model based on a recursive autoencoder, and training the feature extraction model; constructing a state evaluation model of a gated recurrent neural network considering batch normalization, and training the gated recurrent neural network using a cross entropy loss function through a time-expanded reverse error propagation algorithm; when evaluating the operating status of the transformer, the operating status of the transformer is obtained through the trained feature extraction model and the gated recurrent neural network. Summary of the invention

[0007] In order to solve the technical problems in the prior art, the present invention provides a method for monitoring a transformer working scene of a distribution network. By introducing a distributed identification device including a depth of field extraction component and a state judgment component, the method is used to realize a visual judgment of the touching state of a human head and protective glasses in a transformer working scene used in the distribution network, that is, whether the human body wears protective glasses. Specifically, the average of each depth of field value corresponding to each pixel point of the image area of ​​the human head in a received gradient filtered image is used as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and the average of each depth of field value corresponding to each pixel point of the image area of ​​the protective glasses in the received gradient filtered image is used as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image. When there are pixel points overlapping in position between each pixel point of the image area of ​​the human head in the received gradient filtered image and each pixel point of the image area of ​​the protective glasses in the received gradient filtered image, and the difference between the two overall depth of field values ​​is less than or equal to a set difference limit, it is judged that the human head and the protective glasses are in a touching state and a glasses wearing detection instruction is issued; otherwise, it is judged that the human head and the protective glasses are in a non-touching state and a glasses wearing non-detection instruction is issued.

[0008] According to the present invention, the method comprises:

[0009] Using a connection analysis mechanism to detect whether the currently monitored transformer is connected to the power supply grid, and when it is detected that the currently monitored transformer is connected to the distribution grid, a first detection signal is issued, and when it is detected that the currently monitored transformer is not connected to the distribution grid, a second detection signal is issued;

[0010] The capture processing mechanism is arranged just above the currently monitored transformer, and is used to perform capture processing on the working scene of the transformer when the first detection signal is received, so as to obtain and output a corresponding current capture image;

[0011] Connecting the first-layer conversion device to the snapshot processing mechanism to perform tilt correction processing on the received current snapshot image to obtain and output a corresponding tilt-corrected image;

[0012] Connecting the secondary layer conversion device to the primary layer conversion device to perform image SVD enhancement processing on the received tilt-corrected image to obtain and output a corresponding content-enhanced image;

[0013] Connecting the last layer conversion device to the second layer conversion device to perform gradient sharpening filtering on the received content enhanced image to obtain and output a corresponding gradient filtered image;

[0014] Connecting a distribution identification device to the last layer conversion device and including a depth of field extraction component and a state judgment component, wherein the state judgment component is connected to the depth of field extraction component, wherein the distribution identification device is used to use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image, and when there are overlapping pixel points between the pixel points of the image area of ​​the human head in the received gradient filtered image and the pixel points of the image area of ​​the protective glasses in the received gradient filtered image and the difference between the two overall depth of field values ​​is less than or equal to the set difference limit, it is judged that the human head and the protective glasses are in a touching state and a glasses wearing detection instruction is issued; otherwise, it is judged that the human head and the protective glasses are in a non-touching state and a glasses wearing non-detection instruction is issued;

[0015] The wireless transmission interface is connected to the distributed identification device, so that when the glasses wearing is not detected instruction is received, the glasses wearing is not detected instruction and the transformer number corresponding to the currently monitored transformer are entered into an IP network data packet and then wirelessly transmitted to a remote distribution network monitoring node. DETAILED DESCRIPTION

[0016] An embodiment of the method for monitoring the transformer operation scenario of a power distribution network of the present invention will be described in detail below.

[0017] Example 1

[0018] Embodiment 1 is a method for monitoring a transformer working scene of a power distribution network according to a human head embodiment of the present invention, the method comprising:

[0019] Using a connection analysis mechanism to detect whether the currently monitored transformer is connected to the power supply grid, and when it is detected that the currently monitored transformer is connected to the distribution grid, a first detection signal is issued, and when it is detected that the currently monitored transformer is not connected to the distribution grid, a second detection signal is issued;

[0020] Specifically, using a connection analysis mechanism to detect whether a currently monitored transformer is connected to a power supply grid, and sending a first detection signal when detecting that the currently monitored transformer is connected to a distribution grid, and sending a second detection signal when detecting that the currently monitored transformer is not connected to a distribution grid includes: an FPGA device may be selected to implement the use of a connection analysis mechanism to detect whether a currently monitored transformer is connected to a power supply grid, and sending a first detection signal when detecting that the currently monitored transformer is connected to a distribution grid, and sending a second detection signal when detecting that the currently monitored transformer is not connected to a distribution grid;

[0021] The capture processing mechanism is arranged just above the currently monitored transformer, and is used to perform capture processing on the working scene of the transformer when the first detection signal is received, so as to obtain and output a corresponding current capture image;

[0022] Connecting the first-layer conversion device to the snapshot processing mechanism to perform tilt correction processing on the received current snapshot image to obtain and output a corresponding tilt-corrected image;

[0023] Connecting the secondary layer conversion device to the primary layer conversion device to perform image SVD enhancement processing on the received tilt-corrected image to obtain and output a corresponding content-enhanced image;

[0024] Connecting the last layer conversion device to the second layer conversion device to perform gradient sharpening filtering on the received content enhanced image to obtain and output a corresponding gradient filtered image;

[0025] Connecting a distribution identification device to the last layer conversion device and including a depth of field extraction component and a state judgment component, wherein the state judgment component is connected to the depth of field extraction component, wherein the distribution identification device is used to use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image, and when there are overlapping pixel points between the pixel points of the image area of ​​the human head in the received gradient filtered image and the pixel points of the image area of ​​the protective glasses in the received gradient filtered image and the difference between the two overall depth of field values ​​is less than or equal to the set difference limit, it is judged that the human head and the protective glasses are in a touching state and a glasses wearing detection instruction is issued; otherwise, it is judged that the human head and the protective glasses are in a non-touching state and a glasses wearing non-detection instruction is issued;

[0026] The wireless transmission interface is connected to the distributed identification device, and is used to, when receiving the glasses-wearing-undetected instruction, insert the glasses-wearing-undetected instruction and the transformer number corresponding to the currently monitored transformer into an IP network data packet and then wirelessly transmit it to a remote distribution network monitoring node;

[0027] The wireless transmission interface is further used to temporarily put the glasses wearing non-detection instruction and the transformer number corresponding to the currently monitored transformer into an IP network data packet and then wirelessly transmit it to a remote power distribution network monitoring node when receiving the glasses wearing detection instruction;

[0028] The distribution identification device is used to use the mean of the depth of field values ​​corresponding to the pixel points of the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and use the mean of the depth of field values ​​corresponding to the pixel points of the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image, including: identifying the image area of ​​the human head in the received gradient filtered image based on the distribution range of the brightness value corresponding to the human head, and identifying the image area of ​​the protective glasses in the received gradient filtered image based on the appearance contour features corresponding to the protective glasses;

[0029] The capture processing mechanism is further configured to stop executing the capture processing on the working scene of the transformer when the second detection signal is received.

[0030] Example 2

[0031] Embodiment 2 is a method for monitoring the transformer working scene of a power distribution network shown in an embodiment of the protective glasses of the present invention.

[0032] Different from the first embodiment, the method for monitoring the transformer working scene of the power distribution network in the second embodiment may further include:

[0033] The parameter configuration device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively, so as to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with the parameter configuration services they need respectively;

[0034] Wherein, the parameter configuration device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with the parameter configuration services they need respectively, including: the parameter configuration device uses the IIC configuration bus to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with the parameter configuration services they need respectively;

[0035] The parameter configuration device uses the IIC configuration bus to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with the required parameter configuration services, including: the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device have different IIC configuration addresses respectively;

[0036] The first-layer conversion device, the second-layer conversion device, the last-layer conversion device, and the distribution identification device respectively have different IIC configuration addresses, including: the first-layer conversion device, the second-layer conversion device, the last-layer conversion device, and the distribution identification device respectively have different IIC configuration addresses in a binary value representation mode;

[0037] Among them, the different IIC configuration addresses respectively possessed by the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device are represented in a binary numerical mode, including: the byte lengths of the different IIC configuration addresses respectively possessed by the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device are equal.

[0038] Example 3

[0039] Embodiment 3 is a method for monitoring a transformer operating scenario of a power distribution network according to an embodiment of the present invention.

[0040] Different from the first embodiment, the method for monitoring the transformer working scene of the power distribution network in the third embodiment may further include:

[0041] The positioning operation device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively, so as to provide positioning services to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively;

[0042] Wherein, the positioning operation device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively to provide positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively, including: the positioning operation device uses the GPS positioning mechanism to provide GPS positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively;

[0043] And wherein, the positioning operation device is respectively connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device to provide positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device respectively, including: the positioning operation device adopts the Beidou positioning mechanism to provide Beidou positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device respectively.

[0044] In addition, in the method for monitoring the transformer working scene of the distribution network, the wireless transmission interface is connected to the distributed identification device, and is used to, when receiving the glasses wearing undetected instruction, insert the glasses wearing undetected instruction and the transformer number corresponding to the currently monitored transformer into an IP network data packet and then wirelessly transmit it to a remote distribution network monitoring node, including: the remote distribution network monitoring node is a cloud computing monitoring node, a big data monitoring node or a blockchain monitoring node.

[0045] Technical effects of the present invention:

[0046] (1) Introducing a distributed identification device including a depth of field extraction component and a state judgment component to realize visual judgment of the contact state between the human head and protective glasses at the transformer working site used in the distribution network, that is, whether the human body is wearing protective glasses;

[0047] (2) taking the average of the depth of field values ​​corresponding to the pixel points in the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and taking the average of the depth of field values ​​corresponding to the pixel points in the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image;

[0048] (3) When there are overlapping pixel points between each pixel point of the image area of ​​the human head in the received gradient filtered image and each pixel point of the image area of ​​the protective glasses in the received gradient filtered image and the difference between the two overall depth of field values ​​is less than or equal to the set difference limit, it is judged that the human head and the protective glasses are in a touching state and a glasses wearing detection instruction is issued; otherwise, it is judged that the human head and the protective glasses are in a non-touching state and a glasses wearing non-detection instruction is issued.

[0049] Therefore, the method for monitoring the transformer working scene of the power distribution network of the present invention is compact, safe and reliable. Since the distributed identification device with a customized structure is used to realize the visual judgment of the contact state between the human head and the protective glasses at the working scene of the transformer used in the power distribution network, that is, whether the human body is wearing protective glasses, the management level of the transformer working scene of the power distribution network is improved.

[0050] Although the embodiments of the present invention have been described to some extent with specific examples, it should be understood that the present invention is not limited thereto. It will also be understood by those skilled in the art that various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention.

Claims

1. A method for monitoring a transformer operating scene of a power distribution network, characterized in that: The method comprises: Using a connection analysis mechanism to detect whether the currently monitored transformer is connected to the power supply grid, and when it is detected that the currently monitored transformer is connected to the distribution grid, a first detection signal is issued, and when it is detected that the currently monitored transformer is not connected to the distribution grid, a second detection signal is issued; The capture processing mechanism is arranged just above the currently monitored transformer, and is used to perform capture processing on the working scene of the transformer when the first detection signal is received, so as to obtain and output a corresponding current capture image; Connecting the first-layer conversion device to the snapshot processing mechanism to perform tilt correction processing on the received current snapshot image to obtain and output a corresponding tilt-corrected image; Connecting the secondary layer conversion device to the primary layer conversion device to perform image SVD enhancement processing on the received tilt-corrected image to obtain and output a corresponding content-enhanced image; Connecting the last layer conversion device to the second layer conversion device to perform gradient sharpening filtering on the received content enhanced image to obtain and output a corresponding gradient filtered image; Connecting a distribution identification device to the last layer conversion device and including a depth of field extraction component and a state judgment component, wherein the state judgment component is connected to the depth of field extraction component, wherein the distribution identification device is used to use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and use the average of the depth of field values ​​corresponding to the pixel points of the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image, and when there are overlapping pixel points between the pixel points of the image area of ​​the human head in the received gradient filtered image and the pixel points of the image area of ​​the protective glasses in the received gradient filtered image and the difference between the two overall depth of field values ​​is less than or equal to the set difference limit, it is judged that the human head and the protective glasses are in a touching state and a glasses wearing detection instruction is issued; otherwise, it is judged that the human head and the protective glasses are in a non-touching state and a glasses wearing non-detection instruction is issued; The wireless transmission interface is connected to the distributed identification device, so that when the glasses wearing is not detected instruction is received, the glasses wearing is not detected instruction and the transformer number corresponding to the currently monitored transformer are entered into an IP network data packet and then wirelessly transmitted to a remote distribution network monitoring node.

2. The method for monitoring the transformer working scene of the power distribution network according to claim 1, characterized in that: The wireless transmission interface is also used to temporarily put the glasses wearing non-detection instruction and the transformer number corresponding to the currently monitored transformer into an IP network data packet and then wirelessly transmit it to a remote power distribution network monitoring node when receiving the glasses wearing detection instruction; The distribution identification device is used to use the mean of the depth of field values ​​corresponding to the pixel points of the image area of ​​the human head in the received gradient filtered image as the overall depth of field value of the image area of ​​the human head in the received gradient filtered image, and use the mean of the depth of field values ​​corresponding to the pixel points of the image area of ​​the protective glasses in the received gradient filtered image as the overall depth of field value of the image area of ​​the protective glasses in the received gradient filtered image, including: identifying the image area of ​​the human head in the received gradient filtered image based on the distribution range of the brightness value corresponding to the human head, and identifying the image area of ​​the protective glasses in the received gradient filtered image based on the appearance contour features corresponding to the protective glasses; Wherein, the capture processing mechanism is also used to stop executing the capture processing on the working scene of the transformer when receiving the second detection signal.

3. The method for monitoring the transformer working scene of the power distribution network according to claim 2, characterized in that: The method further comprises: The parameter configuration device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively, so as to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with the parameter configuration services they need respectively; Among them, the parameter configuration device is respectively connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with their respective required parameter configuration services, including: the parameter configuration device uses the IIC configuration bus to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device with their respective required parameter configuration services.

4. The method for monitoring the transformer working scene of the power distribution network according to claim 3, characterized in that: The parameter configuration device uses the IIC configuration bus to provide the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device with their respective required parameter configuration services, including: the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device have different IIC configuration addresses respectively.

5. The method for monitoring the transformer working scene of the power distribution network according to claim 4, characterized in that: The first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device respectively have different IIC configuration addresses, including: the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device respectively have different IIC configuration addresses in binary numerical representation mode.

6. The method for monitoring the transformer working scene of the power distribution network according to claim 5, characterized in that: The different IIC configuration addresses respectively possessed by the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device are represented in binary numerical mode, including: the byte lengths of the different IIC configuration addresses respectively possessed by the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device are equal.

7. The method for monitoring the transformer working scene of the power distribution network according to claim 2, characterized in that: The method further comprises: The positioning operation device is connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively, so as to provide positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device respectively.

8. The method for monitoring the transformer working scene of the power distribution network according to claim 7, characterized in that: The positioning operation device is respectively connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device to provide positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device, including: the positioning operation device adopts the GPS positioning mechanism to provide GPS positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distribution identification device.

9. The method for monitoring the transformer working scene of the power distribution network according to claim 7, characterized in that: The positioning operation device is respectively connected to the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device to provide positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device, including: the positioning operation device adopts the Beidou positioning mechanism to provide Beidou positioning services for the first-layer conversion device, the second-layer conversion device, the last-layer conversion device and the distributed identification device.