Partial discharge detection sensor layout method and system of high-voltage equipment and storage medium

Abstract

The embodiment of the invention provides a partial discharge detection sensor layout method and system of high-voltage equipment and a storage medium. The partial discharge detection sensor layout method of the high-voltage equipment comprises the steps of obtaining type and model information of multiple pieces of high-voltage equipment in a target area; radiation area data and a first installation area of the high-voltage equipment corresponding to the type and model information are determined according to the type and model information and the position information, and the first installation area determines an area where a sensor can be installed according to historical error detection data of the high-voltage equipment; obtaining layout information of a partial discharge detection sensor according to historical defect information and radiation area data corresponding to the high-voltage equipment of the target installation; and arranging a partial discharge detection sensor on the digital twinborn model of the target-installed high-voltage equipment according to the layout information. The accuracy of the layout position of the partial discharge detection sensor can be effectively improved, so that the correctness of partial discharge detection is improved.

Description

Technical Field

[0001] The present invention relates to, but is not limited to, the field of power detection technology, and particularly relates to a method, a system, and a storage medium for arranging partial discharge detection sensors of high-voltage equipment. Background Art

[0002] The impact of partial discharge in high-voltage equipment has the dual characteristics of progressiveness and destructiveness, and comprehensive prevention and control measures such as on-line monitoring technology, regular insulation detection, and environmental control are required. Early detection and handling of problems in partial discharge is the key to avoiding serious consequences, so high requirements are put forward for the correctness of partial discharge detection of high-voltage equipment. At present, the setting of partial discharge sensors for high-voltage equipment is mainly based on the experience of on-site installation engineers. Due to the uneven capabilities of on-site installation engineers and the complex layout of high-voltage equipment on site, the partial discharge sensors set for high-voltage equipment often have problems of misjudgment. Summary of the Invention

[0003] The following is an overview of the subject matter described in detail in this document. This overview is not intended to limit the scope of protection of the claims.

[0004] The main purpose of the embodiments of the present invention is to propose a method, a system, and a storage medium for arranging partial discharge detection sensors of high-voltage equipment, which can improve the accuracy of the layout position of partial discharge detection sensors, thereby improving the correctness of partial discharge detection.

[0005] In a first aspect, an embodiment of the present invention provides a method for arranging partial discharge detection sensors of high-voltage equipment, which is applied to a power grid management system. The power grid management system is provided with digital twin models of multiple high-voltage equipment, and the method includes:

[0006] Obtain the type number information and location information of multiple high-voltage equipment in the target area; According to each type number information and the location information, determine the radiation area data and the first installation area of the high-voltage equipment corresponding to the type number information. The first installation area is an area where sensors can be installed determined according to the historical misdetection data of the high-voltage equipment: Based on the first installation area corresponding to the high-voltage equipment to be installed, obtain the layout information of the partial discharge detection sensors according to the historical defect information corresponding to the high-voltage equipment to be installed and the radiation area data corresponding to other multiple high-voltage equipment that are not to be installed; Arrange and set the partial discharge detection sensors on the digital twin model of the high-voltage equipment to be installed according to the layout information.

[0007] In some alternative embodiments, the layout information of the partial discharge detection sensor is obtained based on the historical defect information corresponding to the high-voltage device to be target-mounted and the radiation area data corresponding to multiple other high-voltage devices not to be target-mounted in the first installation area corresponding to the high-voltage device to be target-mounted, including: Input the first installation area, the historical defect information, and the radiation area data into a trained layout model of the partial discharge detection sensor for multi-modal candidate scoring processing, and output the layout information of the partial discharge detection sensor. The layout model of the partial discharge detection sensor includes an input encoding layer, a multi-modal fusion layer, and a spatial optimization layer.

[0008] In some alternative embodiments, the step of inputting the first installation area, the historical defect information, and the radiation area data into a trained layout model of the partial discharge detection sensor for multi-modal candidate scoring processing and outputting the layout information of the partial discharge detection sensor includes: Input the first installation area, the historical defect information, and the radiation area data into the input encoding layer, and respectively perform encoding processing on the first installation area, the historical defect information, and the radiation area data through the input encoding layer to obtain a region feature map, a defect feature vector, and a radiation feature vector; Perform cross-modal attention fusion processing on the region feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer to obtain a fusion feature map; Perform candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor.

[0009] In some alternative embodiments, the spatial optimization layer includes a candidate point generation sub-layer, a layout scoring sub-layer, and a constraint optimization sub-layer. The step of performing candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor includes: Perform candidate processing on the first installation area through the candidate point generation sub-layer to obtain candidate point coordinates; Perform feature extraction and scoring processing on the fusion feature map and the candidate point coordinates through the layout scoring sub-layer to obtain candidate point position scores and sensor type probabilities; Perform radiation constraint processing on the candidate point coordinates, the candidate point position scores, and the sensor type probabilities through the constraint optimization sub-layer to obtain the layout information of the partial discharge detection sensor.

[0010] In some alternative embodiments, the layout information of the partial discharge detection sensor is obtained based on the historical defect information corresponding to the high-voltage device to be target-mounted and the radiation area data corresponding to multiple other high-voltage devices not to be target-mounted, and includes: Determine the non-installation area in the digital twin model of the high-voltage device to be target-mounted according to the radiation area data of multiple high-voltage devices not to be target-mounted; Perform a removal process on the first installation area according to the non-installation area to obtain a second installation area; Determine the layout information of the partial discharge detection sensor according to the second installation area and the historical defect information, and the layout information of the partial discharge detection sensor includes the type and quantity of the partial discharge detection sensor.

[0011] In some alternative embodiments, the method for generating the first installation area includes: Obtain the historical false detection data corresponding to the high-voltage device, where the historical false detection data includes the sensor position information set by the partial discharge detection sensor on the high-voltage device and the information on the cause of the historical recorded false detection corresponding to the partial discharge detection sensor; Determine the first installation area on the digital twin model of the high-voltage device to be target-mounted according to the sensor position information, the information on the cause of the false detection, and the preset installation position information.

[0012] In some alternative embodiments, determining the first installation area on the digital twin model of the high-voltage device to be target-mounted according to the sensor position information, the information on the cause of the false detection, and the preset installation position information includes: When the information on the cause of the false detection is that the first data detected by the partial discharge detection sensor corresponding to the sensor position information does not meet the requirements of partial discharge positioning calculation, and the second data detected by other partial discharge detection sensors meets the requirements of partial discharge positioning calculation, perform area optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be target-mounted; Or, When the information on the cause of the false detection is that there are errors in the third data detected multiple times by the same partial discharge detection sensor corresponding to the sensor position information, perform area optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be target-mounted; Or, In the case where the misdetection cause information is that in the sensor position information, there are multiple different types of partial discharge detection sensors, and the detected fourth data does not meet the requirements of partial discharge positioning calculation, perform regional optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be installed.

[0013] In some alternative embodiments, the determining the radiation area data and the first installation area of the high-voltage device corresponding to each type number information according to the type number information and the position information includes: Search according to each type number information in a preset high-voltage device radiation data table to determine the radiation range data of the high-voltage device corresponding to the type number information. The high-voltage device radiation data table is obtained by detecting the electromagnetic field radiation situation of the high-voltage device, and the high-voltage device radiation data table includes power frequency electromagnetic field data, radio frequency radiation data, and thermal radiation data; Determine the radiation area data of the high-voltage device corresponding to the type number information according to the radiation range data and the position information; Search according to each type number information in the device information table to determine the first installation area corresponding to the type number information.

[0014] In some alternative embodiments, the training method of the partial discharge detection sensor layout model includes: Obtain a sensor layout training text and the corresponding sensor layout true label of the sensor layout training text. The sensor layout training text includes the first installation area, the historical defect information, and the radiation area data; Input the sensor layout training text into an initial partial discharge detection sensor layout model. The partial discharge detection sensor layout model includes an input encoding layer, a multi-modal fusion layer, and a spatial optimization layer; Respectively perform encoding processing on the first installation area, the historical defect information, and the radiation area data through the input encoding layer to obtain a region feature map, a defect feature vector, and a radiation feature vector; Perform cross-modal attention fusion processing on the region feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer to obtain a fusion feature map; Perform candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor; Based on the layout information of the partial discharge detection sensors and the true labels of the sensor layout, calculate the multi-task loss value to obtain the total multi-task loss value, and based on the total multi-task loss value, iteratively update the partial discharge detection sensor layout model to obtain the trained partial discharge detection sensor layout model.

[0015] In some optional embodiments, the spatial optimization layer includes a candidate point generation sub-layer, a layout evaluation sub-layer, and a constraint optimization sub-layer. The candidate scoring process of the fused feature map and the first installation area by the spatial optimization layer to obtain the layout information of the partial discharge detection sensors includes: Perform candidate processing on the first installation area through the candidate point generation sub-layer to obtain candidate point coordinates; Perform feature extraction and scoring processing on the fused feature map and the candidate point coordinates through the layout evaluation sub-layer to obtain candidate point position scores and sensor type probabilities; Perform radiation constraint processing on the candidate point coordinates, the candidate point position scores, and the sensor type probabilities through the constraint optimization sub-layer to obtain the layout information of the partial discharge detection sensors. The layout information of the partial discharge detection sensors includes sensor position information, sensor type information, and sensor position score information; The calculation of the multi-task loss value based on the layout information of the partial discharge detection sensors and the true labels of the sensor layout to obtain the total multi-task loss value includes: Calculate the loss value between the sensor position information and the region feature map to obtain the sensor position loss value; Calculate the type loss value by calculating the loss value between the sensor type information and the true sensor position information in the true labels of the sensor layout; Perform radiation constraint loss calculation on the sensor position score information and the radiation feature vector to obtain the position score loss value; Perform total loss calculation on the sensor position loss value, the sensor type loss value, and the position score loss value to obtain the total multi-task loss value.

[0016] In some optional embodiments, the determination of the second installation area according to the first installation area corresponding to the high-voltage equipment to be target-installed and the other multiple radiation area data of the high-voltage equipment that is not target-installed includes: Render the first installation area corresponding to the high-voltage equipment to be target-installed, and display the first installation area in the digital twin model of the high-voltage equipment to be target-installed; Render the other multiple radiation area data of the high-voltage equipment for non-target installation, and display the radiation areas on the digital twin models of the high-voltage equipment for multiple target installations respectively; Determine the overlapping parts of the multiple radiation areas and the first installation area as non-installation areas; Perform a removal process on the first installation area according to the non-installation areas to obtain a second installation area.

[0017] In a second aspect, an embodiment of the present invention provides a controller, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the partial discharge detection sensor layout method for the high-voltage equipment described in the first aspect.

[0018] In a third aspect, an embodiment of the present invention provides a power grid management system, including the controller described in the second aspect above.

[0019] In a fourth aspect, a computer storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the partial discharge detection sensor layout method for the high-voltage equipment described in the first aspect.

[0020] The beneficial effects of the present invention include: by obtaining the type number information and location information of multiple high-voltage devices in the target area to determine the layout of the high-voltage devices in the target area, the radiation situation between the high-voltage devices can be determined subsequently based on the layout of the high-voltage devices. Then, according to each type number information and location information, the radiation area data and the first installation area corresponding to the high-voltage device of the type number information are determined. The first installation area is the area where sensors can be installed determined according to the historical false detection data of the high-voltage device. The radiation area data and the data of the first installation area of all the high-voltage devices in the target area are obtained for subsequent preparation for calculating the layout information of the partial discharge detection sensors of the target high-voltage device. Since the first installation area is determined by the historical false detection data, some location information that is not suitable for installing a certain type of sensor can be excluded. Then, when the user determines that a new or modified high-voltage device needs to be detected for partial discharge, and the layout calculation of the partial discharge detection sensor for this high-voltage device is required, at this time, the system, based on the first installation area corresponding to the high-voltage device to be installed, according to the historical defect information corresponding to the high-voltage device to be installed and the radiation area data corresponding to other multiple non-target installed high-voltage devices, obtains the layout information of the partial discharge detection sensors, and then arranges and sets the partial discharge detection sensors on the digital twin model of the high-voltage device to be installed according to the layout information. The embodiment of this technical solution initially locates the installation position of the high-voltage device with the target installed sensor through the historical false detection data to obtain the first installation area, and then performs a secondary positioning process on the radiation area data brought by the high-voltage devices near the high-voltage device with the target installed sensor and the historical defect information of this target high-voltage device, and finally determines the layout information of the partial discharge detection sensors, that is, the layout information of the partial discharge detection sensors determined by fully considering the historical situation of the high-voltage device with the target installed sensor itself and the radiation situation of the high-voltage devices near it can effectively improve the accuracy of the layout position of the partial discharge detection sensors, thereby improving the correctness of the partial discharge detection.

[0021] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the specification, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is the step flow block diagram of a method for arranging partial discharge detection sensors of a high-voltage device provided by an embodiment of the present invention; Figure 2 is the schematic diagram of the partial discharge detection sensor layout model provided by an embodiment of the present invention; Figure 3 is the schematic diagram of the training method of the partial discharge detection sensor layout model provided by an embodiment of the present invention. Figure 4 It is a schematic diagram of a controller provided by an embodiment of the present invention. Detailed implementation manners

[0023] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0024] It should be noted that although functional module division is performed in the device schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the sequence in the flowchart. Terms such as "first", "second", etc. in the specification, claims or the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0025] The influence of partial discharge in high-voltage equipment has the dual characteristics of gradualness and destructiveness, and comprehensive prevention and control measures such as on-line monitoring technology, regular insulation detection, and environmental control are required. Early detection and treatment of problems in partial discharge are the key to avoiding serious consequences, so high requirements are put forward for the correctness of partial discharge detection of high-voltage equipment. At present, the setting of partial discharge sensors for high-voltage equipment is mainly based on the experience of on-site installation engineers. Due to the uneven capabilities of on-site installation engineers and the complex layout of on-site high-voltage equipment, the partial discharge sensors set for high-voltage equipment often have problems of misjudgment.

[0026] To solve the above existing problems, the present application provides a method, a system and a storage medium for arranging partial discharge detection sensors of high-voltage equipment.

[0027] In the present application, a method, a system and a storage medium for arranging partial discharge detection sensors of high-voltage equipment are provided, and will be described in detail one by one in the following embodiments.

[0028] Such as Figure 1As shown in the figure, an embodiment of the present invention provides a method for arranging partial discharge detection sensors of high-voltage equipment, which is applied to a power grid management system. The power grid management system is provided with a digital twin model of high-voltage equipment, and a partial discharge detection sensor layout interface with multiple partial discharge sensors arranged on the digital twin model of high-voltage equipment. Through the partial discharge detection sensor layout interface, comprehensive judgment can be made based on the radiation situation of other high-voltage equipment received by the target high-voltage equipment in the environment and the detection situation of the historical sensor layout of the target high-voltage equipment, so as to generate a relatively reasonable sensor layout scheme, and arrange multiple partial discharge sensors at different positions on the digital twin model of high-voltage equipment to improve the accuracy of the layout position of the partial discharge detection sensors, thereby improving the correctness of partial discharge detection.

[0029] It should be noted that the partial discharge sensor can be arranged inside or outside the digital twin model of the high-voltage equipment, and this embodiment does not make specific limitations on it.

[0030] It should be noted that the multiple partial discharge sensors can be evenly distributed or distributed in areas according to the actual situation of the layout requirements, and this embodiment does not make specific limitations on it.

[0031] Specifically, the characteristics of the digital twin model of high-voltage equipment can include: the model is a 1:1 three-dimensional modeling of the appearance of high-voltage equipment and can be displayed in three dimensions in software; for any coordinate point in the three-dimensional scene where the model is located, the model can independently judge whether the point is inside or outside the high-voltage equipment; for any coordinate point inside the high-voltage equipment, the model can independently judge which high-voltage equipment component is inside the high-voltage equipment at this position. The modeling method can include: using a 3D modeling tool (such as Solidworks), according to the drawings of high-voltage equipment or the dimensions measured on site, perform three-dimensional modeling on the high-voltage equipment. When modeling, model the box body, outer surface of the pipeline or the shell of the high-voltage equipment with a thickness of 0.001 meters, and also need to model the inside of the pipeline. Digitalize the space of the three-dimensional model of the high-voltage equipment.

[0032] It should be noted that the high-voltage equipment can be a gas-insulated switchgear, a transformer, or a high-voltage distribution cabinet, etc., and this embodiment does not make specific limitations on it.

[0033] The partial discharge monitoring of high-voltage equipment can be based on the intelligent synchronous acquisition of multiple sensors for the partial discharge monitoring and automatic positioning technology of transformers. By integrating the data of sensors such as ultra-high frequency, high frequency, and ultrasonic wave, constructing a digital twin model of the transformer, and building a fingerprint library of the time difference of arrival of multiple sensors, it is possible to achieve the combined synchronous monitoring of electricity-electricity, sound-electricity, and sound-sound, and optimize the partial discharge positioning results. Due to the strengthening of the algorithm capabilities for partial discharge monitoring and judgment, the sensor layout of high-voltage equipment can be diversified according to the actual situation, not limited to the layout of pure ultrasonic sensors or the layout of pure high-frequency / ultra-high frequency sensors.

[0034] It should be noted that regarding the synchronous monitoring technology of ultra-high frequency, high frequency, and ultrasonic wave, it mainly uses synchronous sampling technology to assign a synchronous clock with an accuracy better than nanoseconds to each ultra-high frequency, high frequency, and ultrasonic wave monitoring channel inside the monitoring device, so that the signal waveforms detected by different monitoring channels have synchronous timestamps, thus having the ability to accurately compare the arrival times of waveforms and meeting the prerequisite conditions for precise positioning; then, by collecting the different types of partial discharge pattern data detected by different types of sensors, obtaining the data characteristics of the abnormal signals in different frequency bands generated by the partial discharges of various internal defects of transformers, and performing multi-source fusion recognition of the partial discharge types; using the fully automatic intelligent positioning algorithms based on the three principles of ultra-high frequency "electricity-electricity" positioning, ultrasonic wave and high frequency "sound-electricity positioning", and ultrasonic wave "sound-sound positioning", when the multi-source sensors detect the partial discharge signal, it is possible to perform online automatic positioning simultaneously based on multiple principles. Through the positioning accuracies of the three principles of ultra-high frequency "electricity-electricity" positioning, ultrasonic wave and high frequency "sound-electricity positioning", and ultrasonic wave "sound-sound positioning" for different defect types, when the defect type is successfully identified, it is also possible to automatically determine which principle's positioning result should be preferentially used as the final positioning result.

[0035] Applied to the above power grid management system, the method for arranging the partial discharge detection sensors of the high-voltage equipment includes: S100. Obtain the type number information and location information of multiple high-voltage equipment in the target area.

[0036] Specifically, in the same distribution substation in the power grid management system, there will be various multiple high-voltage equipment, such as transformers, high-voltage switchgear cabinets, high-voltage switches, and gas-insulated switchgear, etc. Then, for the update, replacement, or addition of a certain high-voltage equipment, the radiation area of the newly added or changed high-voltage equipment changes, which may affect the sensor detection of other high-voltage equipment in the distribution substation. Therefore, it is necessary to first obtain the type number information of multiple high-voltage equipment in the target area. And due to the different structures in the distribution substation, the types and locations of the high-voltage equipment arranged are different. Therefore, it is necessary to obtain the type number information and location information of multiple high-voltage equipment in the target area near the high-voltage equipment where the sensor is to be installed or changed.

[0037] S200. Determine the radiation area data and the first installation area of the high-voltage equipment corresponding to each type number information and position information, where the first installation area is the area where sensors can be installed determined according to the historical misdetection data of the high-voltage equipment.

[0038] Specifically, the radiation range data of high-voltage equipment of different type numbers are all different. The radiation range data of high-voltage equipment are obtained by engineering personnel or through a measurement robot detecting the radiation range data during the operation of the high-voltage equipment. In the power grid management system, there is a data table corresponding to the radiation range data, and this data table corresponding to the radiation range data is preset in the power grid management system. In addition, in the power grid management system, there is also an equipment information table for various high-voltage equipment, and the first installation area information for suggesting the installation of partial discharge detection sensors is also set in this information table. Then, combined with the position information of each high-voltage equipment, the radiation area data of a certain high-voltage equipment relative to the radiation area data of other high-voltage equipment can be obtained. After setting up the digital twin models of multiple high-voltage equipment, according to the radiation range data, a radiation circular image can be generated on the digital twin model of the high-voltage equipment corresponding to the radiation range data and fused with the digital twin model of the high-voltage equipment. Then, for other high-voltage equipment located within the radiation range of the radiation circular image of the digital twin model of the high-voltage equipment, the radiation range data of this high-voltage equipment to other high-voltage equipment can be obtained.

[0039] It should be noted that for a position point of a high-voltage equipment that is within the radiation ranges of multiple other high-voltage equipment, then this position point can be the radiation area data generated by the superposition state of the radiation range data of multiple other high-voltage equipment, or the highest value among the radiation range data of multiple other high-voltage equipment can be taken as the radiation area data, and this embodiment does not make specific limitations on it.

[0040] It should be noted that different measuring instruments can be selected according to different radiation types. For example, for power frequency electromagnetic field measurement, an electric field intensity meter for measuring electric field intensity or a magnetic field intensity meter for measuring magnetic induction intensity can be used; for another example, for radio frequency radiation measurement, a frequency and intensity spectrum analyzer for analyzing high-frequency electromagnetic waves or a radio frequency field strength meter for measuring radio frequency radiation power density can be used; for another example, for thermal radiation measurement, an infrared thermal imager for detecting the surface temperature distribution of the device or a temperature sensor for contact temperature measurement can be used.

[0041] In some alternative embodiments, step S200 may include looking up from a preset high-voltage equipment radiation data table according to each type number information to determine the radiation range data of the high-voltage equipment corresponding to the type number information. The high-voltage equipment radiation data table is obtained by detecting the electromagnetic field radiation condition of the high-voltage equipment. The high-voltage equipment radiation data table includes power frequency electromagnetic field data, radio frequency radiation data, and thermal radiation data. Then, according to the radiation range data and the position information, determine the radiation area data of the high-voltage equipment corresponding to the type number information; look up from the equipment information table according to each type number information to determine the first installation area corresponding to the type number information.

[0042] Among them, the data of the first installation area may also be, based on the suggestions in the equipment information table, the data of misdetection that appears in the data detected by the partial discharge detection sensor during the daily operation of a large number of high-voltage equipment, that is, historical misdetection data. The data of the first installation area can also be further corrected according to the frequency and deviation of the occurrence of such data to obtain the data of the new first installation area, and update the data in the equipment information table.

[0043] It should be noted that the method for generating the first installation area may include: obtaining the historical misdetection data corresponding to the high-voltage equipment. The historical misdetection data includes the sensor position information set by the partial discharge detection sensor on the high-voltage equipment and the misdetection reason information corresponding to the historical record of the partial discharge detection sensor. Then, determine the first installation area according to the sensor position information, the misdetection reason information, and the position information of the high-voltage equipment.

[0044] In some alternative embodiments, when performing distance conversion on the partial discharge signals of the high-voltage equipment obtained by multiple different types or the same type of partial discharge detection sensors, if the deviation in the process of positioning judgment between the distance of the partial discharge signal calculated by a certain partial discharge detection sensor from this partial discharge detection sensor and the distance of the partial discharge signals calculated by other partial discharge detection sensors from this partial discharge detection sensor exceeds 20%, then the data detected by this partial discharge detection sensor this time can be determined as misdetection data and recorded; then, in the case that this partial discharge detection sensor still continues to have misdetection data, the area covered by this partial discharge detection sensor and the nearby areas can be defined as the area to be corrected and excluded from the original first installation area.

[0045] It should be noted that because the anti-interference ability and the ability to monitor partial discharge signals of different types of partial discharge detection sensors are different, the first installation areas corresponding to different types of partial discharge detection sensors may be different. Of course, considering the complexity of data storage and data operation, the first installation areas corresponding to different types of partial discharge detection sensors can also be set to be the same.

[0046] In some alternative embodiments, when the cause information of the misdetection is that the first data detected by the partial discharge detection sensor corresponding to the sensor position information does not meet the requirements of the partial discharge positioning calculation, and the second data detected by other partial discharge detection sensors meets the requirements of the partial discharge positioning calculation, regional optimization processing is performed according to the preset installation position information and the sensor position information to determine the first installation area.

[0047] In some alternative embodiments, when the cause information of the misdetection is that the third data detected by the same partial discharge detection sensor corresponding to the sensor position information is incorrect for multiple detections, regional optimization processing is performed according to the preset installation position information and the sensor position information to determine the first installation area.

[0048] In some alternative embodiments, when the cause information of the misdetection is that, among the sensor position information, the fourth data detected by multiple different types of partial discharge detection sensors does not meet the requirements of the partial discharge positioning calculation, regional optimization processing is performed according to the preset installation position information and the sensor position information to determine the first installation area.

[0049] S300. Based on the first installation area corresponding to the high-voltage equipment to be target-installed, the layout information of the partial discharge detection sensors is obtained according to the historical defect information corresponding to the high-voltage equipment to be target-installed and the radiation area data corresponding to multiple other high-voltage equipment that are not target-installed.

[0050] Specifically, if the first installation area corresponding to the high-voltage equipment to be target-installed is the position information where sensors are recommended or can be installed, then the layout design problem of the sensors existing in some of the self-high-voltage equipment with historical sensor misdetections can be excluded through the first installation area. Then, based on the historical defect information corresponding to the high-voltage equipment to be target-installed and the radiation area data brought by other high-voltage equipment near the installation site, the specific situation on-site is further restricted, thereby effectively improving the accuracy of the layout information of the partial discharge detection sensors.

[0051] In some alternative embodiments, the non-installation area is determined in the digital twin model of the high-voltage equipment to be target-installed according to the radiation area data of multiple high-voltage equipment that are not target-installed; then, the first installation area is removed according to the non-installation area to obtain the second installation area; and then, the layout information of the partial discharge detection sensors is determined according to the second installation area and the historical defect information. The layout information of the partial discharge detection sensors includes the types and quantities of the partial discharge detection sensors.

[0052] Specifically, based on the radiation area data of multiple non-target installed high-voltage devices and combining the position information of the non-target installed high-voltage devices and the target installed high-voltage devices, the non-installation area can be determined in the digital twin model of the target installed high-voltage device. The non-installation area is the area with relatively large device radiation from the non-target installed high-voltage devices. It can be understood that the radiation area data can be used to decompose the area according to the radiation height in the digital twin model of the non-target installed high-voltage device and display it in a circular shape. For example, the high-radiation area is the red ring area near the center of the non-target installed high-voltage device, the medium-radiation area is the yellow ring area outside the high-radiation area, and the low-radiation area is the green ring area outside the medium-radiation area. Different types of partial discharge detection sensors have different anti-interference capabilities, and the acceptable radiation areas corresponding to different types of partial discharge detection sensors are different. Then, the area that coincides with the radiation area in the digital twin model of the non-target installed high-voltage device in the first installation area can be determined as the radiation situation received by the partial discharge detection sensor. Thus, the non-installation area can be determined according to the type of partial discharge detection sensor required for this area. Then, the first installation area is processed by removing the non-installation area, that is, the non-installation area is removed to obtain the second installation area. Then, the layout information of the partial discharge detection sensor is determined according to the second installation area and the historical defect information. The layout information of the partial discharge detection sensor includes the type and quantity of the partial discharge detection sensor. For the second installation area where the distance of the historical defect information meets the first distance threshold, the quantity or type of the partial discharge detection sensor can be appropriately increased to solve the problem of inaccuracy caused by radiation.

[0053] In some alternative embodiments, first render the first installation area corresponding to the target installed high-voltage device and display the first installation area in the digital twin model of the target installed high-voltage device; then render the radiation area data of other multiple non-target installed high-voltage devices and display the radiation areas in the digital twin models of multiple target installed high-voltage devices respectively; then determine the overlapping part of the multiple radiation areas and the first installation area as the non-installation area; process the first installation area by removing the non-installation area to obtain the second installation area.

[0054] In some alternative embodiments, refer to Figure 2, input the first installation area, historical defect information, and radiation area data into the trained partial discharge detection sensor layout model for multi-modal candidate scoring processing, and output the layout information of the partial discharge detection sensor. The partial discharge detection sensor layout model includes an input encoding layer 210, a multi-modal fusion layer 220, and a spatial optimization layer 230. Among them, the historical defect information is the defect information caused by aging or damage existing in the target high-voltage equipment. The defect information existing in the high-voltage equipment will be reflected in the form of partial discharge, that is, the position where partial discharge occurs is associated with the position corresponding to the defect information, and the position correlations corresponding to different types of defect information are different.

[0055] Specifically, input the first installation area, historical defect information, and radiation area data into the trained partial discharge detection sensor layout model for multi-modal candidate scoring processing, and output the layout information of the partial discharge detection sensor, including: first input the first installation area, historical defect information, and radiation area data into the input encoding layer 210, and perform encoding processing on the first installation area, historical defect information, and radiation area data respectively through the input encoding layer 210 to obtain a region feature map, a defect feature vector, and a radiation feature vector; then perform cross-modal attention fusion processing on the region feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer 220 to obtain a fusion feature map; and then perform candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer 230 to obtain the layout information of the partial discharge detection sensor.

[0056] It should be noted that the partial discharge detection sensor layout model in this embodiment is a multi-modal graph neural network, and the input data includes the first installation area, historical defect information, and radiation area data. Among them, the historical defect information is represented by a two-dimensional heat map, which is used to represent the probability or frequency of defects occurring at each position of the equipment. The first installation area; the radiation area data is represented in a graph structure, the nodes include the target high-voltage equipment and other surrounding high-voltage equipment, and the edges represent radiation effects with intensity weights. In addition, the radiation area corresponding to each radiation source device can be represented by polygon coordinates, but when inputting into the model, it can be converted into a radiation heat map with the same size as the defect heat map of the defect information, and each pixel value represents the radiation intensity at that position, or directly use the graph structure data, node features, and adjacency matrix; the first installation area is usually a binary mask with a size similar to the defect heat map of the historical defect information, where 1 represents installable and 0 represents non-installable.

[0057] The input encoding layer 210 includes three encoders, namely, a defect thermal map encoder 211, a radiation map encoder 212, and an installation area encoder 213. Among them, the defect thermal map encoder 211 uses a convolutional neural network to extract features from historical defect information. By setting several convolutional layers and pooling layers, and finally flattening or using global average pooling, a defect feature vector of historical defect information is obtained. The radiation map encoder 212 superimposes the radiation areas of high-voltage equipment of each radiation source to generate a radiation thermal map, and then uses a convolutional neural network to extract a radiation feature vector of the radiation area data. The installation area encoder 213 also uses a convolutional neural network to extract features to obtain a regional feature map. The multi-modal fusion layer 220 is used to fuse the regional feature map, the defect feature vector, and the radiation feature vector. First, the regional feature map, the defect feature vector, and the radiation feature vector are concatenated, and then through a fully connected layer, the concatenated data is dimensionally reduced or feature-integrated to obtain a fused feature map. The spatial optimization layer 230 may include a candidate point generation sub-layer 231, a layout scoring sub-layer 232, and a constraint optimization sub-layer 233. Then, through the spatial optimization layer 230, candidate scoring processing is performed on the fused feature map and the first installation area to obtain the layout information of the partial discharge detection sensor, including: first, the candidate point generation sub-layer 231 performs candidate processing on the first installation area to obtain candidate point coordinates; then, the layout scoring sub-layer 232 performs feature extraction and scoring processing on the fused feature map and the candidate point coordinates to obtain candidate point position scores and sensor type probabilities; finally, the constraint optimization sub-layer 233 performs radiation constraint processing on the candidate point coordinates, the candidate point position scores, and the sensor type probabilities to obtain the layout information of the partial discharge detection sensor. The neural network model of the technical solution of this embodiment realizes an anti-interference optimized layout in a complex multi-device environment by fusing the radiation influence relationship and historical defect conditions between devices, and at the same time satisfies the engineering installation constraints, effectively improving the accuracy of the layout position of the partial discharge detection sensor, thereby improving the correctness of partial discharge detection.

[0058] In some alternative embodiments, the partial discharge detection sensor layout model includes an input encoding layer 210, a multi-modal fusion layer 220, and a spatial optimization layer 230. The spatial optimization layer 230 includes a candidate point generation sub-layer 231, a layout scoring sub-layer 232, and a constraint optimization sub-layer 233. The input data of the partial discharge detection sensor layout model: Defect thermal map: Matrix [H, W], Radiation data: Can be a thermal map [H, W] or a graph structure (node feature matrix and adjacency matrix), Installable area: Binary mask [H, W]. The output data is the sensor layout information: A list containing K elements, each element being a tuple (x, y, sensor_type), which represents the type and layout position information of each sensor.

[0059] The model data processing of the partial discharge detection sensor layout model is as follows: Defect heat map -> CNN encoder -> Defect feature vector; Radiation data -> Graph encoder -> Radiation feature vector; Installation area -> CNN encoder -> Installable feature vector. Then, the three feature vectors -> Multimodal fusion layer 220 (concatenation) -> Fusion feature vector; Fusion feature vector + Candidate point position encoding -> Fully connected network (shared weights) -> Score and type probability of each candidate point; Then select the candidate point with the highest score (considering constraints) -> Output sensor layout information.

[0060] It should be noted that considering that in actual engineering, the installable area may already contain information such as safety distance, so when generating candidate points, it can be ensured that the distance between them is greater than the minimum installation distance. In this way, distance constraints do not need to be considered during scoring, and only sorting by score and selection are required during the final selection because the candidate points already meet the minimum distance.

[0061] It should be noted that before inputting into the model, candidate points can be generated within the installable area. The generation method can be uniform grid sampling (for example, one point every 0.1 meters), or adaptive sampling based on the shape of the installable area (dense at the edges and sparse in the middle). After generating the candidate points, their coordinates can be converted into position encoding (for example, normalized to 0 - 1 and then embedded as a vector).

[0062] In some alternative embodiments, Historical defect information [H, W] -> CNN (convolutional layer, pooling layer) -> Global average pooling -> Defect feature vector [D]; Radiation area data: (Node feature matrix [N, F], Adjacency matrix [N, N]) -> GCN (2 layers) -> Target device node embedding [D]; First installation area [H, W] -> CNN (a few simple layers) -> Radiation feature vector [D]; Concatenation of the three feature vectors [3 * D] -> Fully connected layer (dimensionality reduction to D) -> Fusion feature map [D]; For each candidate point: Candidate point coordinates (x, y) -> Position encoding (fully connected layer) -> Position feature vector [D]; Concatenation of the fusion feature vector [D] and the position feature vector [D] -> [2 * D] -> Fully connected layer -> Output: Installation probability (0 to 1) and type probability (4 classes); Then sort the installation probabilities of each candidate point and select the top K points with the highest probability.

[0063] It should be noted that the model outputs the installation probability and type probability of each candidate point. A probability threshold (such as 0.5) is set to determine whether to install, but this may exceed K. Therefore, the K points with the highest probabilities can be selected (K is the preset number of sensors). If it is less than K, all are selected; if it exceeds K, the first K are selected. Then, check whether these K points meet the minimum distance constraint (if not, remove the points with lower probabilities until it is satisfied). Finally, output the positions and types of these points.

[0064] In some alternative embodiments, referring to Figure 3 , a method for training a partial discharge detection sensor layout model includes: Step S310: Obtain a sensor layout training text and a sensor layout true label corresponding to the sensor layout training text. The sensor layout training text includes a first installation area, historical defect information, and radiation area data; Step S320: Input the sensor layout training text into an initial partial discharge detection sensor layout model. The partial discharge detection sensor layout model includes an input encoding layer, a multi-modal fusion layer, and a spatial optimization layer; Step S321: Respectively perform encoding processing on the first installation area, historical defect information, and radiation area data through the input encoding layer to obtain an area feature map, a defect feature vector, and a radiation feature vector; Step S322: Perform cross-modal attention fusion processing on the area feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer to obtain a fusion feature map; Step S323: Perform candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor; Step S330: Calculate a multi-task total loss value based on the layout information of the partial discharge detection sensor and the sensor layout true label, and iteratively update the partial discharge detection sensor layout model based on the multi-task total loss value to obtain a trained partial discharge detection sensor layout model.

[0065] Specifically, data preparation is carried out first: mainly prepare three types of input data (the first installation area, historical defect information, and radiation area data) and label data (sensor layout positions and types). The three types of input data are respectively encoded by three encoders (defect heat map encoder, radiation map encoder, installation area encoder), and then multi-modal fusion is performed on the three types of encoded area feature maps, defect feature vectors, and radiation feature vectors. Then, the layout information of the predicted partial discharge detection sensor is obtained through the spatial optimization layer. Next, the loss calculation of the layout information of the partial discharge detection sensor is carried out, including calculating the position loss, type loss, and radiation constraint loss. According to the loss calculation results, backpropagation and optimization are performed, and the weights are updated using gradient descent, that is, the layout model of the partial discharge detection sensor is iteratively updated based on the multi-task total loss value to obtain a trained layout model of the partial discharge detection sensor.

[0066] It should be noted that during training, the label data is prepared, that is, the optimal sensor layout (position and type) of each high-voltage device. The label data can be a manually designed layout or a layout obtained through simulation. The form of the label data is: for each candidate point, there is a binary label (whether to install a sensor) and a sensor type label (if it is a positive sample). Alternatively, the layout problem can also be transformed into a multi-label classification problem (predicting whether to install and the installed type for each candidate point). Among them, for the setting of the loss function, since each candidate point is independent, the binary cross-entropy loss (whether to install) and the multi-class cross-entropy loss (sensor type) can be used for each candidate point. It can be set that only K sensors are required, so the ranking loss (such as TopK loss) or the global loss (such as set prediction loss) can be used. However, for simplicity, the installation decision for each candidate point can also be regarded as a binary classification, and the cross-entropy loss with weight adjustment is used, which is applicable to the case of few positive samples. For type prediction, the cross-entropy loss can be calculated only for positive samples (installation points). In addition, a constraint violation penalty term (such as the penalty for two sensors being too close) can be added.

[0067] It should be noted that each candidate point in the training data has a label (0 or 1, indicating whether to install) and a sensor type (if installed). Therefore, two losses can be defined: installation loss: binary cross-entropy (or focal loss); type loss: multi-class cross-entropy (calculated only for installation points). And the total loss = installation loss + α * type loss. Since the number of sensors K is fixed in the actual layout, K can also be used as an input to control the number when the model outputs. But usually K is predetermined, so K can be fixed.

[0068] S400. Layout and set the partial discharge detection sensors on the digital twin model of the high-voltage device to be installed according to the layout information.

[0069] Specifically, according to the layout information, partial discharge detection sensors are arranged on the digital twin model of the target installed high-voltage equipment. After the arrangement, the display interface can show the target installed high-voltage equipment, the partial discharge detection sensors arranged on the high-voltage equipment, and the radiation range map of the non-installed high-voltage equipment. The partial discharge detection sensors arranged at positions near the boundary between two radiation levels are marked, and the changed or newly added partial discharge detection sensors can also be marked, so that the installation decision-maker can intuitively see the situation of the changed partial discharge detection sensors, facilitating decision-making and modification.

[0070] It should be noted that the type and anti-radiation ability data of the sensor can also be marked on the marked partial discharge detection sensor, and other data can also be added. The data can be displayed constantly or displayed after selecting the icon corresponding to the partial discharge detection sensor. This embodiment does not make specific limitations on this.

[0071] The present invention determines the layout of high-voltage devices in a target area by obtaining the type numbers and location information of multiple high-voltage devices in the target area. Subsequently, the radiation situation between high-voltage devices can be determined based on the layout of high-voltage devices. Then, according to each type number information and location information, the radiation area data and the first installation area corresponding to the high-voltage device corresponding to the type number information are determined. The first installation area is the area where sensors can be installed determined according to the historical misdetection data of the high-voltage device. The radiation area data of high-voltage devices in all target areas and the data of the first installation area are obtained to prepare for subsequent calculation of the layout information of partial discharge detection sensors for target high-voltage devices. Since the first installation area is determined by historical misdetection data, some location information that is not suitable for installing a certain type of sensor can be excluded. Then, when the user determines that a new or modified high-voltage device needs to be installed and requires layout calculation of partial discharge detection sensors for this high-voltage device, the system, based on the first installation area corresponding to the target-installed high-voltage device, obtains the layout information of partial discharge detection sensors according to the historical defect information corresponding to the target-installed high-voltage device and the radiation area data corresponding to other multiple non-target-installed high-voltage devices. Then, according to the layout information, partial discharge detection sensors are arranged and set on the digital twin model of the target-installed high-voltage device. In the embodiment of this technical solution, the installation position of the high-voltage device where the target sensor is installed is initially located through historical misdetection data to obtain the first installation area. Then, through secondary positioning processing of the radiation area data brought by high-voltage devices near the high-voltage device where the target sensor is installed and the historical defect information of this target high-voltage device, the layout information of partial discharge detection sensors is finally determined, that is, the layout information of partial discharge detection sensors determined by fully considering the historical situation of the high-voltage device where the target sensor is installed and the radiation situation of high-voltage devices near it can effectively improve the accuracy of the layout position of partial discharge detection sensors, thereby improving the correctness of partial discharge detection.

[0072] As shown in FIG. 4, Figure 4 A structural block diagram of a controller 1000 provided according to an embodiment of the present application is shown. The components of the controller 1000 include, but are not limited to, a memory 1200 and a processor 1100. The processor 1100 is connected to the memory 1200 through a bus, and the memory 1200 is used to store data.

[0073] The controller 1000 also includes an access device that enables the controller 1000 to communicate via one or more networks. Examples of such networks include the Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device may include one or more of any type of wired or wireless network interface (e.g., Network Interface Card (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, Worldwide Interoperability for Microwave Access (Wi-MAX) interface, Ethernet interface, Universal Serial Bus (USB) interface, cellular network interface, Bluetooth interface, Near Field Communication (NFC) interface, and so on.

[0074] The controller 1000 can be any type of stationary or mobile electronic device, including mobile computers or mobile electronic devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable electronic devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary electronic devices such as desktop computers or PCs. The controller 1000 of the intelligent seat can also be a mobile or stationary server.

[0075] Wherein, the processor 1100 is used to execute the computer-executable instructions of the partial discharge detection sensor layout method for high-voltage equipment.

[0076] The above is a schematic solution of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the partial discharge detection sensor layout method for high-voltage equipment belong to the same concept. For the details not described in detail in the technical solution of the computing device, reference can be made to the description of the technical solution of the partial discharge detection sensor layout method for high-voltage equipment.

[0077] According to an embodiment of the present application, there is also provided a power grid management system. The power grid management system is installed with the controller 1000, or the power grid management system is communicatively connected to the controller 1000, so that the power grid management system realizes the positioning of the partial discharge source through the controller 1000. It should be noted that the technical solution of this computing device and the technical solution of the partial discharge detection sensor layout method for high-voltage equipment belong to the same concept. For the details not described in detail in the technical solution of the computing device, reference can be made to the description of the technical solution of the partial discharge detection sensor layout method for high-voltage equipment.

[0078] The embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, it implements the partial discharge detection sensor layout method for high-voltage equipment.

[0079] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the objectives of the solution of this embodiment.

[0080] Those of ordinary skill in the art can understand that all or some of the steps and systems disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium generally includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.

[0081] The above has specifically described the preferred embodiments of the present application, but the present application is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present application, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present application.

Claims

1. A method for arranging partial discharge detection sensors of high-voltage equipment, characterized in that, Applied to a power grid management system, the power grid management system is provided with digital twin models of multiple high-voltage devices, and the method includes: Obtain the type number information and location information of multiple high-voltage devices in the target area; Determine the radiation area data and the first installation area corresponding to the high-voltage device corresponding to the type number information according to each type number information and the location information, and the first installation area is the area where sensors can be installed determined according to the historical misdetection data of the high-voltage device: Based on the first installation area corresponding to the high-voltage device to be installed, obtain the layout information of the partial discharge detection sensor according to the historical defect information corresponding to the high-voltage device to be installed and the radiation area data corresponding to other multiple high-voltage devices that are not to be installed; Layout and set the partial discharge detection sensor on the digital twin model of the high-voltage device to be installed according to the layout information.

2. The partial discharge detection sensor layout method for high-voltage equipment according to claim 1, wherein The step of obtaining the layout information of the partial discharge detection sensor based on the first installation area corresponding to the high-voltage device to be installed, according to the historical defect information corresponding to the high-voltage device to be installed and the radiation area data corresponding to other multiple high-voltage devices that are not to be installed, includes: Input the first installation area, the historical defect information, and the radiation area data into a trained partial discharge detection sensor layout model for multi-modal candidate scoring processing, and output the layout information of the partial discharge detection sensor. The partial discharge detection sensor layout model includes an input encoding layer, a multi-modal fusion layer, and a spatial optimization layer.

3. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 2, characterized in that The step of inputting the first installation area, the historical defect information, and the radiation area data into a trained partial discharge detection sensor layout model for multi-modal candidate scoring processing, and outputting the layout information of the partial discharge detection sensor, includes: Input the first installation area, the historical defect information, and the radiation area data into the input encoding layer, and respectively perform encoding processing on the first installation area, the historical defect information, and the radiation area data through the input encoding layer to obtain a region feature map, a defect feature vector, and a radiation feature vector; Perform cross-modal attention fusion processing on the region feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer to obtain a fusion feature map; Perform candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor.

4. The method for arranging partial discharge detection sensors of high-voltage equipment according to claim 3, characterized in that, The spatial optimization layer includes a candidate point generation sub-layer, a layout scoring sub-layer, and a constraint optimization sub-layer. The step of performing candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor includes: Perform candidate processing on the first installation area through the candidate point generation sub-layer to obtain candidate point coordinates; Perform feature extraction and scoring processing on the fusion feature map and the candidate point coordinates through the layout scoring sub-layer to obtain candidate point position scores and sensor type probabilities; The radiation constraint processing is performed on the candidate point coordinates, the candidate point position scores, and the sensor type probabilities through the constraint optimization sublayer to obtain the layout information of the partial discharge detection sensors.

5. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 1, characterized in that Based on the first installation area corresponding to the high-voltage device to be target-mounted, the layout information of the partial discharge detection sensors is obtained according to the historical defect information corresponding to the high-voltage device to be target-mounted and the radiation area data corresponding to multiple other high-voltage devices to be non-target-mounted, including: Determine the non-installation area in the digital twin model of the high-voltage device to be target-mounted according to the radiation area data of the multiple high-voltage devices to be non-target-mounted; Perform removal processing on the first installation area according to the non-installation area to obtain a second installation area; Determine the layout information of the partial discharge detection sensors according to the second installation area and the historical defect information, and the layout information of the partial discharge detection sensors includes the types and quantities of the partial discharge detection sensors.

6. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 1, characterized in that The method for generating the first installation area includes: Obtain the historical false detection data corresponding to the high-voltage device, where the historical false detection data includes the sensor position information set by the partial discharge detection sensor on the high-voltage device and the information on the reasons for the historical recorded false detections corresponding to the partial discharge detection sensor; Determine the first installation area on the digital twin model of the high-voltage device to be target-mounted according to the sensor position information, the reasons for false detections information, and the preset installation position information.

7. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 6, characterized in that The determining of the first installation area on the digital twin model of the high-voltage device to be target-mounted according to the sensor position information, the reasons for false detections information, and the preset installation position information includes: In the case where the information on the reasons for false detections is that the first data detected by the partial discharge detection sensor corresponding to the sensor position information does not meet the requirements of partial discharge positioning calculation, and the second data detected by other partial discharge detection sensors meets the requirements of partial discharge positioning calculation, perform area optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be target-mounted; Or, In the case where the information on the reasons for false detections is that there are errors in the third data detected multiple times by the same partial discharge detection sensor corresponding to the sensor position information, perform area optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be target-mounted; Or, In the case where the information on the reasons for false detections is that among the sensor position information, there are multiple different types of partial discharge detection sensors whose detected fourth data do not meet the requirements of partial discharge positioning calculation, perform area optimization processing according to the preset installation position information and the sensor position information, and determine the first installation area on the digital twin model of the high-voltage device to be target-mounted.

8. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 1, characterized in that Determining the radiation area data and the first installation area of the high-voltage equipment corresponding to each of the type model information and the location information includes: Searching according to each of the type model information in a preset high-voltage equipment radiation data table to determine the radiation range data of the high-voltage equipment corresponding to the type model information. The high-voltage equipment radiation data table is obtained by detecting the electromagnetic field radiation situation of the high-voltage equipment, and the high-voltage equipment radiation data table includes power frequency electromagnetic field data, radio frequency radiation data, and thermal radiation data; Determining the radiation area data of the high-voltage equipment corresponding to the type model information according to the radiation range data and the location information; Searching according to each of the type model information in the equipment information table to determine the first installation area corresponding to the type model information.

9. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 2, characterized in that, The training method of the partial discharge detection sensor layout model includes: Obtaining a sensor layout training text and a sensor layout true label corresponding to the sensor layout training text. The sensor layout training text includes the first installation area, the historical defect information, and the radiation area data; Inputting the sensor layout training text into an initial partial discharge detection sensor layout model. The partial discharge detection sensor layout model includes an input encoding layer, a multi-modal fusion layer, and a spatial optimization layer; Respectively performing encoding processing on the first installation area, the historical defect information, and the radiation area data through the input encoding layer to obtain an area feature map, a defect feature vector, and a radiation feature vector; Performing cross-modal attention fusion processing on the area feature map, the defect feature vector, and the radiation feature vector through the multi-modal fusion layer to obtain a fusion feature map; Performing candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor; Calculating a multi-task total loss value based on the layout information of the partial discharge detection sensor and the sensor layout true label, and iteratively updating the partial discharge detection sensor layout model based on the multi-task total loss value to obtain the trained partial discharge detection sensor layout model.

10. The method for arranging partial discharge detection sensors of a high-voltage device according to claim 9, characterized in that, The spatial optimization layer includes a candidate point generation sub-layer, a layout scoring sub-layer, and a constraint optimization sub-layer. The performing candidate scoring processing on the fusion feature map and the first installation area through the spatial optimization layer to obtain the layout information of the partial discharge detection sensor includes: Performing candidate processing on the first installation area through the candidate point generation sub-layer to obtain candidate point coordinates; Performing feature extraction and scoring processing on the fusion feature map and the candidate point coordinates through the layout scoring sub-layer to obtain candidate point position scores and sensor type probabilities; Radiation constraint processing is performed on the candidate point coordinates, the candidate point position scores, and the sensor type probabilities through the constraint optimization sublayer to obtain the layout information of the partial discharge detection sensors. The layout information of the partial discharge detection sensors includes sensor position information, sensor type information, and sensor position score information; Calculating the multi-task total loss value based on the layout information of the partial discharge detection sensors and the true label of the sensor layout includes: Calculating the loss value between the sensor position information and the region feature map to obtain the sensor position loss value; Calculating the type loss value between the sensor type information and the true sensor position information in the true label of the sensor layout to obtain the sensor type loss value; Performing radiation constraint loss calculation on the sensor position score information and the radiation feature vector to obtain the position score loss value; Calculating the total loss of the sensor position loss value, the sensor type loss value, and the position score loss value to obtain the multi-task total loss value.

11. A power grid management system, characterized in that, Including a controller, the controller includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the method for arranging partial discharge detection sensors of high-voltage equipment according to any one of claims 1-10.

12. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions for executing the method for arranging partial discharge detection sensors of high-voltage equipment according to any one of claims 1-10.

Images