A method and system for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging

Abstract

This application provides a method and system for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging, belonging to the field of power equipment condition monitoring technology. This application first acquires ultraviolet photon and spatial point cloud data collected by a UAV. A three-dimensional model is constructed based on the point cloud, and electric field distribution information is obtained through finite element analysis. Combining photon data and electric field information, the discharge charge quantity and distribution are obtained through spatial inversion correction. Then, the discharge current density is calculated using an energy dissipation model, and the discharge heat power loss density is determined by combining the local electric field. Finally, the steady-state heat conduction equation is solved based on heat conduction parameters to calculate the temperature field distribution under thermal equilibrium. When the steady-state temperature exceeds the heat resistance threshold, an early warning message is generated. This application achieves early and accurate prediction and location of hidden overheating faults in electrical equipment caused by insulation degradation.

Description

Technical Field

[0001] This application belongs to the field of power equipment condition monitoring technology, and in particular relates to a method and system for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging. Background Technology

[0002] With the advancement of smart grid construction, using drones equipped with multispectral imaging devices for non-contact inspection of power transmission and transformation equipment has become an important means of power operation and maintenance. This technology can simultaneously acquire multidimensional information such as visible light, infrared, and ultraviolet light, and has broad application prospects in improving inspection efficiency and ensuring the safe and stable operation of the power grid.

[0003] Existing inspection methods based on UAV multispectral imaging typically utilize infrared thermal imaging technology to directly measure the surface temperature of equipment to detect obvious thermal defects, or use ultraviolet imaging technology to count photons to qualitatively determine whether external corona discharge exists. Some existing technologies attempt to perform simple visual overlay of infrared and ultraviolet images to help maintenance personnel visually view the positional relationship between discharge points and heat sources.

[0004] However, most existing detection methods only represent surface temperature or discharge signals independently, lacking quantitative analysis of the physical mechanism by which the electrical energy generated by the discharge is converted into thermal energy. This makes it impossible to establish an accurate mapping model from weak partial discharge to internal temperature rise in the insulating medium. Consequently, when early discharge occurs inside the equipment due to insulation degradation but no significant temperature rise has yet appeared on the surface, it is difficult to effectively predict its potential overheating trend. Therefore, existing technologies suffer from untimely and inaccurate early warning systems for hidden overheating faults in electrical equipment caused by early partial discharge. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging, so as to solve the problem of insufficient early warning of overheating hazards in electrical equipment in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a method for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging, comprising:

[0007] Acquire photon density data and spatial point cloud data in the ultraviolet band from the surface of electrical equipment using a drone equipped with a multispectral imaging device;

[0008] A three-dimensional geometric model is constructed based on spatial point cloud data to determine the physical interface of electrical equipment. Finite element analysis is then performed on the physical interface and the operating voltage level of the electrical equipment to obtain electric field distribution information, including local electric field intensity, electric field vector direction and equipotential surface information.

[0009] Based on photon density data and electric field distribution information, the discharge charge of the discharge channel and the corresponding spatial distribution data are obtained by mapping and correction using the spatial inversion of the local discharge charge.

[0010] Based on the discharge charge, spatial distribution data, and electric field distribution information at the corresponding locations, the discharge current density in the discharge channel is calculated using the constructed discharge energy dissipation model. Based on the discharge current density and local electric field strength, the density distribution data of the discharge heat power loss of the electrical equipment insulation layer is determined.

[0011] Based on density distribution data and the thermal conductivity and heat dissipation parameters of electrical equipment, the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation is calculated by solving the steady-state heat conduction equation. The steady-state temperature of different areas of the electrical equipment is determined, and an early warning message is generated when there is an area where the steady-state temperature is greater than the preset heat resistance threshold.

[0012] Optionally, a three-dimensional geometric model is constructed based on spatial point cloud data to determine the physical interfaces of the electrical equipment, including:

[0013] Based on the spatial coordinate adjacency relationship, adjacent coordinate points in the spatial point cloud data are topologically connected to construct a triangular mesh surface covering the outer contour of the electrical equipment;

[0014] By closing the triangular mesh surfaces, a three-dimensional geometric model representing the surface topology of electrical equipment is generated.

[0015] Based on the geometric morphological characteristics of the three-dimensional geometric model, the solid region is extracted from the three-dimensional geometric model, and the interface between the solid region and the external space is taken as the physical interface.

[0016] Optionally, by performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment, electric field distribution information, including local electric field intensity, electric field vector direction, and equipotential surface information, can be obtained, including:

[0017] Based on the physical interface, the three-dimensional geometric model is divided into an insulating region and an air region, and the corresponding dielectric parameters are determined for the insulating region and the air region respectively.

[0018] Based on the operating voltage level and dielectric parameters, determine the potential boundary conditions of the insulation region to construct the potential distribution equation;

[0019] By iteratively solving the potential distribution equation, the potential value of each spatial node is obtained, and spatial nodes with the same potential value are used as equipotential surface information.

[0020] By calculating the spatial gradient of the potential value, the electric field vector direction and local electric field intensity corresponding to each spatial node are obtained, and all local electric field intensities, all electric field vector directions and equipotential surface information are combined into electric field distribution information.

[0021] Optionally, based on photon density data and electric field distribution information, the discharge charge of the discharge channel and its corresponding spatial distribution data are obtained by mapping and correction using spatial inversion of the local discharge charge, including:

[0022] Based on the distribution characteristics of local electric field intensity and electric field vector direction on the surface of the three-dimensional geometric model, the surface region that meets the gas ionization condition is extracted from the surface of the three-dimensional geometric model, and the three-dimensional coordinate set of the surface region is determined as spatial distribution data;

[0023] The photon density data is mapped to the spatial location corresponding to the spatial distribution data. The optical path distance from each spatial location to the UAV is calculated to obtain the corresponding attenuation factor. The attenuation factor is used to correct the mapped photon density data to obtain the total number of photons.

[0024] The total number of photons is converted into the discharge charge of the discharge channel using a preset photoelectric conversion coefficient, which is determined by the gas discharge luminescence mechanism.

[0025] Optionally, based on the discharge charge quantity, spatial distribution data, and corresponding electric field distribution information, the discharge current density within the discharge channel is calculated using the constructed discharge energy dissipation model, including:

[0026] The connected regions in the spatial distribution data where the local electric field strength is greater than the preset breakdown threshold are identified as active discharge regions. The maximum span of the active discharge region is calculated along the electric field vector direction to obtain the path length of the discharge channel.

[0027] Calculate the average value of the projected area of ​​the cross-section of the active discharge region to obtain the cross-sectional area of ​​the discharge channel;

[0028] The equivalent current of the discharge channel is obtained by calculating the ratio of the discharge charge to the electron drift time corresponding to the path length.

[0029] The ratio of the equivalent current to the cross-sectional area is calculated to obtain the scalar current density of the discharge channel. The direction of the scalar current density is set to be consistent with the direction of the electric field vector to obtain the discharge current density.

[0030] Optionally, based on the discharge current density and local electric field strength, the density distribution data of the discharge heat power loss of the electrical equipment insulation layer is determined, including:

[0031] The thermal power density at each spatial location is obtained by performing a dot product operation on the discharge current density and the local electric field intensity.

[0032] The loss density generated by the local electric field intensity under the action of an alternating electric field is obtained by multiplying the square of the modulus of the local electric field intensity by the dielectric loss factor of the insulating layer.

[0033] The thermal power density and loss density are linearly superimposed and mapped to the corresponding spatial nodes in the three-dimensional geometric model to obtain density distribution data.

[0034] Optionally, the thermal conductivity and heat dissipation parameters include the material's specific heat capacity, thermal conductivity, and heat dissipation coefficient;

[0035] Based on density distribution data and the thermal conductivity and dissipation parameters of electrical equipment, the temperature field distribution at which internal heat generation and surface heat dissipation reach thermal equilibrium is calculated by solving the steady-state heat conduction equation. This determines the steady-state temperature of different regions of the electrical equipment, including:

[0036] Density distribution data is loaded as a heat source term into the insulating region of the three-dimensional geometric model, and the specific heat capacity and thermal conductivity of the material are assigned to the insulating region to obtain the insulating entity data;

[0037] The heat dissipation coefficient is applied as a heat transfer boundary condition to the physical interface of the three-dimensional geometric model to obtain boundary condition data.

[0038] Based on the data of the insulating entity and the boundary conditions, a steady-state heat conduction equation is constructed according to the law of conservation of energy. The temperature value of each spatial node in the three-dimensional geometric model is obtained by iteratively solving the steady-state heat conduction equation.

[0039] All temperature values ​​are mapped to their corresponding spatial locations in a three-dimensional geometric model to obtain the temperature field distribution. The local temperature sets corresponding to different areas of the electrical equipment are then extracted from the temperature field distribution as steady-state temperatures.

[0040] Secondly, this application provides an electrical equipment overheating hazard early warning system based on UAV multispectral imaging, comprising:

[0041] The acquisition module is used to acquire photon density data and spatial point cloud data in the ultraviolet band of the surface of electrical equipment collected by a drone equipped with a multispectral imaging device.

[0042] The generation module is used to construct a three-dimensional geometric model based on spatial point cloud data to determine the physical interface of electrical equipment, and to obtain electric field distribution information including local electric field intensity, electric field vector direction and equipotential surface information by performing finite element analysis on the physical interface and the operating voltage level of electrical equipment.

[0043] The generation module is also used to obtain the discharge charge and corresponding spatial distribution data of the discharge channel by mapping and correcting based on photon density data and electric field distribution information through spatial inversion of local discharge charge.

[0044] The calculation module is used to calculate the discharge current density in the discharge channel based on the discharge charge, spatial distribution data and electric field distribution information at the corresponding location, using the constructed discharge energy dissipation model. Based on the discharge current density and local electric field strength, the density distribution data of the discharge heat power loss of the electrical equipment insulation layer is determined.

[0045] The solution module is used to calculate the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation based on density distribution data and the thermal conductivity and heat dissipation parameters of electrical equipment by solving the steady-state heat conduction equation. It determines the steady-state temperature of different areas of the electrical equipment and generates an early warning message when there is an area where the steady-state temperature is greater than the preset heat resistance threshold.

[0046] Thirdly, this application provides an electronic device, comprising:

[0047] Memory, used to store computer programs;

[0048] A processor is configured to execute the computer program to implement the steps of the method for early warning of electrical equipment overheating hazards based on UAV multispectral imaging as described in the first aspect above.

[0049] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the electrical equipment overheating hazard early warning method based on UAV multispectral imaging as described in the first aspect above.

[0050] The method for early warning of overheating hazards of electrical equipment based on UAV multispectral imaging provided in this application integrates ultraviolet photon data with three-dimensional electric field distribution information constructed from spatial point clouds, and uses spatial inversion and correction techniques to convert qualitative light signals into quantitative discharge charge, effectively overcoming the defect that relying solely on photon counting cannot accurately assess the discharge intensity.

[0051] Furthermore, this application innovatively introduces a discharge energy dissipation model, and based on the coupled calculation of discharge current density and local electric field strength, constructs a physical mapping relationship from microscopic charge migration to macroscopic thermal power loss, filling the gap in the analysis of the mechanism of discharge electrical energy to thermal energy conversion.

[0052] Based on this, by solving the steady-state heat conduction equation using the equipment's thermal conductivity and heat dissipation parameters, the temperature field distribution inside the insulating medium under thermal equilibrium can be accurately predicted. This allows for the early identification of hidden overheating risks caused by early partial discharges, even before significant temperature rises are observed on the equipment surface. Therefore, this application effectively solves the technical problems of untimely and inaccurate early warning systems for hidden overheating faults in electrical equipment caused by early partial discharges in existing technologies. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating an electrical equipment overheating hazard early warning method based on UAV multispectral imaging, provided in an embodiment of this application;

[0055] Figure 2 A flowchart illustrating a method for generating discharge charge quantity and spatial distribution data provided in an embodiment of this application;

[0056] Figure 3 A flowchart illustrating a method for determining steady-state temperature provided in an embodiment of this application;

[0057] Figure 4 A schematic diagram of an electrical equipment overheating hazard early warning system based on UAV multispectral imaging provided in this application embodiment;

[0058] Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0059] Although UAV-based multispectral imaging technology for electrical equipment inspection has been widely used in power operation and maintenance, existing infrared thermometry and ultraviolet imaging methods mainly rely on the direct capture of abnormal surface temperatures of equipment or simple statistics of the number of discharge photons.

[0060] This superficial detection method cannot grasp the essence of energy conversion behind the discharge phenomenon. Especially in the early stage of insulation deterioration, although weak partial discharge has occurred, it has not yet caused a significant increase in the temperature of the equipment surface. Existing technology cannot establish a quantitative connection between electrical signals and thermal effects, which often leads to delayed or even missed early warning of hidden overheating faults, seriously restricting the safety of power grid operation.

[0061] To address the aforementioned issues, this application proposes a method for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging. The core of this method lies in constructing a physical mapping model from microscopic discharge signals to macroscopic thermal field distribution. Specifically, this application utilizes ultraviolet photon density and spatial point cloud data synchronously collected by a UAV, combined with electric field distribution information obtained from finite element analysis, to accurately quantify the discharge charge through spatial inversion technology. Furthermore, a discharge energy dissipation model is introduced to calculate the heat power loss generated by the coupling of discharge current density and local electric field, and the final temperature field distribution inside and on the surface of the equipment is deduced based on the steady-state heat conduction equation.

[0062] This method breaks through the limitations of traditional single signal detection. Through deep fusion and calculation of multi-physics field data, it can quantify and predict the overheating trend caused by discharge in advance when the temperature rise on the equipment surface is not obvious. It solves the problems of untimely warning and low accuracy of existing technologies for early hidden faults, and significantly improves the foresight and reliability of electrical equipment condition assessment.

[0063] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] To address the problems of existing technologies, embodiments of this application provide a method, apparatus, device, computer storage medium, and computer program product for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging. The method for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging provided in this application embodiment will be described below first.

[0065] Figure 1 This illustration shows a flowchart of an embodiment of a method for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging. Figure 1 As shown, the method includes:

[0066] S101. Acquire photon density data and spatial point cloud data in the ultraviolet band on the surface of electrical equipment using a drone equipped with a multispectral imaging device.

[0067] Photon density data in the ultraviolet band refers to the statistical value of the number of discharge photons received per unit area on the surface of a target electrical device per unit time by a solar-blind ultraviolet imaging module carried by a drone. It can reflect the intensity distribution of corona discharge or partial discharge on the surface of the device.

[0068] Spatial point cloud data refers to a set of discrete coordinate points describing the spatial geometric position of the surface of electrical equipment, collected by a lidar or binocular vision camera mounted on a drone. Each point includes the three-dimensional coordinate information of the equipment surface. A multispectral imaging device is a comprehensive acquisition device integrating multiple sensors such as visible light, infrared thermal imaging, and ultraviolet imaging, capable of simultaneously acquiring spectral information of the same target in different wavelength bands. Furthermore, this device or drone platform is also equipped with micro-meteorological sensors to simultaneously collect environmental parameters such as ambient temperature, humidity, and wind speed.

[0069] In practice, the first step is to control a drone equipped with a multispectral imaging device to fly along a preset route to the vicinity of target electrical equipment, such as high-voltage composite insulator strings or transformer bushings, maintaining a hovering or low-speed cruising state. Next, a solar-blind ultraviolet camera is used to photograph the equipment, collecting radiation signals in the ultraviolet band. The processor inside the imaging device converts the received light signals into a digitized photon counting matrix, generating photon density data. ,in Indicates the first in the image Line number Photon count value of each pixel.

[0070] Simultaneously, the lidar onboard the drone scans the electrical equipment, emitting laser beams and receiving echoes to calculate the three-dimensional spatial coordinates of the equipment's surface and generate spatial point cloud data. ,in This represents the total number of point clouds. Simultaneously, an airborne micro-weather sensor was used to record the current ambient temperature in real time. relative humidity and wind speed These environmental parameters are then associated with and stored in conjunction with the collected photon data and point cloud data.

[0071] Finally, the collected photon density data D, spatial point cloud data p, and environmental parameters are time-stamped and synchronized, and then transmitted back to the ground station or stored in the airborne computing unit in real time via a wireless transmission link.

[0072] S102. Construct a three-dimensional geometric model based on spatial point cloud data to determine the physical interface of the electrical equipment, and obtain electric field distribution information including local electric field intensity, electric field vector direction and equipotential surface information by performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment.

[0073] Optionally, the process of constructing a three-dimensional geometric model based on spatial point cloud data to determine the physical interface of the electrical equipment in step S102 may specifically include:

[0074] S1021. Connect adjacent coordinate points in the spatial point cloud data according to the spatial coordinate adjacency relationship to construct a triangular mesh surface covering the outer contour of the electrical equipment.

[0075] A triangular mesh is a set of triangular patches formed by connecting three non-collinear point cloud coordinates in space. It is a common geometric representation for transitioning discrete point cloud data to continuous curved surfaces and can approximate the local shape of electrical equipment surfaces. Topological connection refers to the process of determining which points should be connected into a triangle based on the spatial positional relationships of point cloud data, such as Euclidean distance. It is usually implemented using Delaunay triangulation or greedy projection triangulation algorithms.

[0076] In practice, the spatial point cloud data p obtained from S101 is first imported, and then denoised and downsampled to optimize data quality. Next, a greedy projection triangulation algorithm is used to search for each point. of In the nearest neighbor region, based on the conditions of normal vector consistency and maximum angle constraint, three coordinate points that satisfy the adjacency relationship are connected to form a triangular patch. ,in Repeat this process until all point cloud data has been traversed, ultimately generating a set of triangular mesh surfaces covering the entire outer contour of the electrical equipment. ,in This represents the total number of triangular facets.

[0077] S1022. By closing the triangular mesh surface, a three-dimensional geometric model representing the surface topology of the electrical equipment is generated.

[0078] Closure processing refers to the process of repairing holes, gaps, or non-manifold structures such as overlapping surfaces and isolated points in a triangular mesh to make it a mathematically closed and leak-free geometry. A three-dimensional geometric model is a digital model that, after closure processing, can fully express the geometry, volume, and surface topology of electrical equipment, and is usually stored in STL or OBJ format.

[0079] In practice, the triangular mesh face set generated by S1021 is first detected. The system checks for boundary edges, i.e., edges that belong to only one triangle. These boundary edges typically correspond to holes caused by missing data. Next, using wavefront or radial basis function interpolation algorithms, new triangular patches are automatically generated to fill these hole regions, and these new patches are incorporated into the mesh set.

[0080] Simultaneously, self-intersecting patches and non-manifold edges in the model are checked and removed to ensure the geometric topological correctness of the model. Finally, the mesh surface is fine-tuned using the Laplacian smoothing algorithm to eliminate local sharp protrusions caused by noise, generating a smooth and closed 3D geometric model. .

[0081] S1023. Based on the geometric morphological characteristics of the three-dimensional geometric model, extract the solid region from the three-dimensional geometric model, and take the interface between the solid region and the external space as the physical interface.

[0082] The solid region refers to the spatial volume enclosed within the three-dimensional geometric model, representing the insulating medium of the electrical equipment itself, such as ceramics, silicone rubber, or composite materials. The physical interface is the boundary surface between the solid region and the external air domain, i.e., the outer surface of the electrical equipment; it is the critical boundary where the dielectric constant changes abruptly in electric field analysis. Geometric morphological characteristics refer to the geometric properties of the model surface, such as curvature, normal vector direction, and connectivity.

[0083] In practice, the three-dimensional geometric model is first analyzed. Based on the surface normal vector direction and the principle that the normal vector points outward, the interior and exterior sides of the model are determined. Next, using voxelization algorithms or tetrahedral meshing techniques, the interior space of the model is filled with solid elements, defined as solid regions. Based on the specific material type of the electrical equipment, such as a high-voltage bushing, the corresponding relative permittivity is retrieved from the material property database. to give the region Each finite element mesh element in the process.

[0084] At the same time, construct a geometric model that surrounds the entire structure. A large-scale bounding box, which includes elements within the bounding box but... The space outside is defined as the air domain. And uniformly assign it the relative permittivity of air. Finally, Boolean operations are used to extract... and The common contact surfaces are marked as physical interfaces. In finite element analysis software, this interface is defined as the junction surface that satisfies the boundary conditions of continuous tangential electric field and discontinuous normal electric displacement vector when electric field lines pass through it. For example, in the simulation model of a high-voltage bushing, the outer surface of the skirt is the physical interface, which separates the internal high-dielectric-constant insulating material from the external air.

[0085] Optionally, step S102, which involves performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment to obtain electric field distribution information including local electric field intensity, electric field vector direction, and equipotential surface information, may specifically include:

[0086] S1024. Divide the three-dimensional geometric model into an insulating region and an air region according to the physical interface, and determine the corresponding dielectric parameters for the insulating region and the air region respectively.

[0087] The insulating region refers to the internal space of the entity surrounded by the physical interface in S1023, representing the main insulating part of the electrical equipment. The air region refers to the space outside the physical interface; typically, in finite element analysis, a finite-sized computational domain, such as a sphere or cube, is set to simulate the infinite boundary. The dielectric parameter mainly refers to the relative permittivity of the material, which determines the distribution and refraction of electric field lines in different media.

[0088] In practice, the physical interface determined in S1023 is first imported. Next, in the preprocessing module of finite element analysis software such as COMSOL or ANSYS, the interface is... The internal geometry is defined as a domain. Based on the equipment ledger information, such as the model B bushing, the relative permittivity of its insulating material, such as epoxy resin, is obtained by referring to a table. Assign it to the field Meanwhile, in the split interface A cubic computing domain with sides 5 to 10 times the device size is constructed on the periphery. The size of this computing domain is then subtracted from the size of the surrounding area. The remaining part is defined as a field. And it assigns the relative permittivity of air to it. .

[0089] S1025. Based on the operating voltage level and dielectric parameters, determine the potential boundary conditions of the insulation region to construct the potential distribution equation.

[0090] Potential boundary conditions refer to known potential values ​​applied at specific boundaries of the finite element computational domain, typically including Dirichlet boundary conditions. Operating voltage levels refer to the rated operating voltage of electrical equipment under actual operating conditions, such as 110kV or 220kV. The potential distribution equation describes the potential in an electrostatic field. Partial differential equations that vary with spatial location are usually Laplace's equations or Poisson's equations.

[0091] In practice, the domain is first identified. High-voltage conductive components such as the surface of the central guide rod and grounding components such as flange base surface Next, the effective value of the current operating phase voltage of the device will be obtained from the SCADA system. Convert to peak Alternatively, an effective value can be taken based on actual calculation needs. In this embodiment, to maintain consistency between the field strength calculation and energy density, the voltage peak value is used. Apply to the boundary That is, setting Simultaneously, zero potential is applied to the boundary. That is, setting For the outer boundary of the air domain, a zero potential or zero charge boundary condition is typically set. Finally, this is combined with the dielectric parameters determined in the previous step. Based on the electrostatic field theory in Maxwell's equations, a potential distribution equation applicable to regions without free charges is constructed: .in, For Hamiltonian operators, The dielectric constant is... China ,exist China , Let be the space potential distribution function to be determined.

[0092] S1026. By iteratively solving the potential distribution equation, the potential value of each spatial node is obtained, and spatial nodes with the same potential value are used as equipotential surface information.

[0093] Iterative solution refers to the process of discretizing continuous partial differential equations into a system of linear algebraic equations using the finite element method, and then gradually approximating the true solution through numerical calculation. Spatial nodes are discrete points generated after finite element mesh generation; they are the basic units for calculating potential values. Equipotential surface information refers to the set of surfaces formed by all points in space with equal potential values, which intuitively reflects the voltage drop trend.

[0094] In practical implementation, the entire computational domain is first... The system of discrete algebraic equations is then divided into millions of tetrahedral mesh elements. Next, the system of equations is submitted to a solver, and iterative calculations are performed using the conjugate gradient method or the multigrid method until the residuals converge to a preset accuracy. After the calculation is complete, extract each grid node. scalar potential value at Finally, an interpolation algorithm is used to search for all values ​​that satisfy the condition. Based on the spatial location, a series of equipotential surfaces are generated, and the geometric data of these surfaces are stored as equipotential surface information. .in For constant series, such as 10kV, 20kV...

[0095] S1027. By calculating the spatial gradient of the potential value, the electric field vector direction and local electric field intensity corresponding to each spatial node are obtained, and all local electric field intensities, all electric field vector directions and equipotential surface information are combined into electric field distribution information.

[0096] Spatial gradient calculation refers to the mathematical process of differentiating a scalar field, i.e., electric potential, to obtain a vector field, i.e., electric field. Local electric field strength refers to the magnitude of the electric field vector at a point in space, representing the strength of the electric force at that point. The direction of the electric field vector indicates the direction in which the electric potential drops the fastest. Electric field distribution information is a comprehensive dataset including all the above-mentioned electrical characteristics.

[0097] In practical implementation, the nodal potential field obtained based on S1026 The negative gradient is calculated using a numerical differentiation algorithm to obtain the electric field intensity vector at each node. As shown in the following formula (1):

[0098] (1)

[0099] in, For Hamiltonian operators, This is a unit direction vector. The calculation results include the electric field vector direction at each node. and local electric field intensity value Ultimately, all nodes... , as well as Packaged and integrated to form complete electric field distribution information .

[0100] This embodiment effectively overcomes the limitation of traditional detection methods in obtaining microscopic electric field information. By utilizing high-precision local electric field strength and vector direction data, it significantly improves the spatial resolution and calculation accuracy of the overheating early warning model.

[0101] S103. Based on photon density data and electric field distribution information, the discharge charge of the discharge channel and the corresponding spatial distribution data are obtained by mapping and correction using the spatial inversion of the partial discharge charge.

[0102] Optionally, step S103, based on photon density data and electric field distribution information, maps and corrects the discharge charge of the discharge channel and the corresponding spatial distribution data by utilizing the spatial inversion of the local discharge charge, and may specifically include:

[0103] Figure 2 A schematic flowchart of a method for generating discharge charge and spatial distribution data according to an embodiment of this application is shown. Figure 2 As shown, the method includes:

[0104] S1031. Based on the distribution characteristics of local electric field intensity and electric field vector direction on the surface of the three-dimensional geometric model, extract the surface region that meets the gas ionization condition from the surface of the three-dimensional geometric model, and determine the three-dimensional coordinate set of the surface region as spatial distribution data.

[0105] Gas ionization conditions refer to the critical electric field strength threshold required for air molecules to undergo electron avalanche and generate self-sustaining discharge under the influence of a strong electric field. For example, the corona discharge field strength of air under standard atmospheric pressure is approximately 30 kV / cm. When the local electric field strength exceeds this threshold, corona discharge is highly probable in that region. Spatial distribution data refers to the set of spatial coordinates of all micro-elements or nodes on the surface of a three-dimensional geometric model that satisfy the above conditions. It precisely defines the possible physical location range of the discharge channel.

[0106] In practice, the electric field distribution information obtained from S1027 is first imported. traverse each spatial node Local electric field intensity value Next, based on Paschen's law and current environmental parameters such as temperature and air pressure, the current critical field strength for air ionization is calculated. .

[0107] By comparing algorithms, all those that satisfy the criteria are selected. The nodes are identified, and the surface region formed by connecting these nodes is marked as the discharge active region. Finally, extract the region. Spatial coordinates of all grid nodes To form a coordinate set This serves as the final spatial distribution data.

[0108] S1032. Map the photon density data to the spatial location corresponding to the spatial distribution data, calculate the optical path distance from each spatial location to the UAV, obtain the corresponding attenuation factor, and use the attenuation factor to correct the mapped photon density data to obtain the total number of photons.

[0109] Optical path distance refers to the straight-line Euclidean distance from the discharge point (i.e., the coordinates in the spatial distribution data) to the center of the imaging plane of the UAV camera. Attenuation factor refers to the proportional coefficient of light intensity loss caused by spherical wave diffusion (geometric attenuation) and atmospheric absorption and scattering (physical attenuation) during ultraviolet photon propagation. Total photon count refers to the actual initial photon emission at the discharge source, calculated after inversion correction.

[0110] In practice, the two-dimensional photon density data D acquired by S101 is first projected onto the three-dimensional spatial position determined by S1031 using the pinhole camera model and the GPS / IMU pose data of the UAV. Above, each discharge node is obtained. Number of observed photons at the location .

[0111] Next, calculate each node. To the location of the drone optical path distance Based on the inverse square law of radiative transfer, the corresponding geometric attenuation factor is calculated. Specifically, only the major geometric diffusion loss is considered here, while weak atmospheric absorption is ignored. Finally, using the formula... The observed data is reverse-corrected, and the total number of photons from the discharge source is obtained by summing the correction values ​​of all nodes. .

[0112] S1033. The total number of photons is converted into the discharge charge of the discharge channel using a preset photoelectric conversion coefficient. The photoelectric conversion coefficient is determined by the gas discharge luminescence mechanism.

[0113] The photoelectric conversion coefficient refers to the ratio of the number of ultraviolet photons generated per unit discharge charge (e.g., 1 picocoulomb, pC) under specific discharge conditions, such as positive polarity streamer discharge. The discharge charge refers to the actual amount of charge transferred within the discharge channel during a partial discharge. Preset photoelectric conversion coefficients are set according to different discharge types, as shown in Table 1 below:

[0114] Table 1: Comparison Table of Photoelectric Conversion Coefficients

[0115]

[0116] As shown in Table 1, Table 1 lists the photon yield efficiencies corresponding to typical corona discharge, streamer discharge, and spark discharge under standard atmospheric conditions.

[0117] In practice, the current discharge type is first identified based on the photon density characteristics collected by S101 or the electric field intensity characteristics extracted by S1031. The specific identification criterion is: the maximum surface electric field intensity calculated based on S1031. ,like If the photon cloud exhibits a sparse distribution, it is determined to be corona discharge; if If the photon cloud exhibits a distinct streamer-like clustered extension, it is determined to be a streamer discharge; if Furthermore, if the ultraviolet signal is accompanied by a sharp, wide-band amplitude jump, it is determined to be a spark discharge. After automatic identification, the corresponding photoelectric conversion coefficient is matched from Table 1. .

[0118] Photoelectric conversion coefficient The construction method is as follows: a standard discharge calibration experimental platform is built, and different types of discharges are generated in a dark room using a standard needle-plate electrode model. The actual discharge pulse current is synchronously collected by a high-frequency current transformer and integrated to obtain the calibration charge. Simultaneously, the multispectral imaging device is used to record the number of photons. The conversion coefficients were obtained through linear regression fitting. Finally, using the formula Calculate the apparent discharge charge of the discharge channel. .

[0119] This embodiment can accurately locate the active discharge area and quantify the apparent discharge charge, effectively solving the limitation that single ultraviolet imaging can only perform qualitative observations, and realizing the quantitative assessment of discharge intensity.

[0120] S104. Based on the discharge charge, spatial distribution data, and electric field distribution information at the corresponding location, the discharge current density in the discharge channel is calculated using the constructed discharge energy dissipation model. Based on the discharge current density and local electric field strength, the density distribution data of the discharge heat power loss of the electrical equipment insulation layer is determined.

[0121] Optionally, step S104, which calculates the discharge current density within the discharge channel using the constructed discharge energy dissipation model based on the discharge charge, spatial distribution data, and corresponding electric field distribution information, may specifically include:

[0122] S1041. Determine the connected regions in the spatial distribution data where the local electric field strength is greater than the preset breakdown threshold as the discharge active region, and calculate the maximum span of the discharge active region along the electric field vector direction to obtain the path length of the discharge channel.

[0123] The preset breakdown threshold refers to the minimum electric field strength required for air to ionize and break down under specific air pressure and temperature. When the local electric field strength exceeds this threshold, air molecules in that region are considered to be ionized, forming a conductive channel. The discharge active region refers to a set of three-dimensional regions that meet the above breakdown conditions and are spatially continuous, representing the actual physical extent of the discharge channel. The path length refers to the maximum geometric distance the discharge channel extends along the direction of the electric field lines, reflecting the longitudinal scale of discharge development. The preset breakdown threshold is set according to different environmental conditions, as shown in Table 2 below:

[0124] Table 2: Breakdown Threshold Comparison Table

[0125]

[0126] As shown in Table 2, Table 2 lists the breakdown field strength after standard atmospheric pressure and correction at different altitudes, which is used to guide the determination of the discharge active region.

[0127] In practice, the environmental parameters recorded by S101, namely air pressure P and temperature, are first used as the basis for implementation. Calculate relative air density ,in The standard atmospheric pressure and temperature are used. Next, the current preset breakdown threshold is calculated using the empirical formula for the gas discharge initiation field strength, namely the Peek formula. The calculation process is shown in the following formula (2):

[0128] (2)

[0129] in, This is the baseline breakdown field strength under standard atmospheric pressure, and its value for air is 30 kV / cm. Relative air density; The radius of curvature of the electrode is determined based on the three-dimensional geometric model; This is the surface roughness coefficient, with a value range of [value range missing]. ; denoted as the electrode geometry coefficient, which is 0.308 in a coaxial structure. This formula accurately describes the physical property that a smaller radius of curvature, i.e., a sharper electrode, results in a larger surface electric field gradient, leading to an increase in the initial discharge field strength.

[0130] Next, the electric field distribution information obtained from S1027 is traversed. Filter out all that satisfy the local electric field intensity The spatial nodes are clustered into several independent connected regions using a connected component labeling algorithm, with each connected region representing a potential active discharge region. .

[0131] Finally, for each Along the electric field vector in this region The average direction is used to calculate its maximum projected length or geodesic distance in space, which is taken as the path length of the discharge channel. .

[0132] S1042. Calculate the average value of the projected area of ​​the cross-section of the active discharge region to obtain the cross-sectional area of ​​the discharge channel.

[0133] The cross-section of the active discharge region refers to the cross-section perpendicular to the direction of the discharge channel extension, i.e., the direction of the electric field vector. The cross-sectional area is a key geometric parameter indicating the thickness of the discharge channel.

[0134] In practice, firstly, along the path length direction determined by S1041, the discharge active region is... Cut into Each layer consists of an equally spaced thin sheet, with a thickness of 1 mm. Next, the projected area of ​​each thin sheet on a plane perpendicular to the electric field vector direction is calculated. Finally, the average projected area of ​​all thin layers is calculated to obtain the average cross-sectional area of ​​the discharge channel. .

[0135] S1043. Calculate the ratio of the discharge charge to the electron drift time corresponding to the path length to obtain the equivalent current of the discharge channel.

[0136] Electron drift time refers to the time required for an electron to traverse the entire path length of a discharge channel under the influence of an electric field. It depends on the electron mobility and the average electric field strength along the path. Equivalent current refers to the macroscopic current intensity generated by the directional movement of charges during a single discharge.

[0137] In practice, the first step is to consider the electron mobility in the air. Electron mobility in air at standard atmospheric pressure and room temperature Approximately Combined with the average electric field strength within the active discharge region. Calculate the average drift velocity of electrons In particular, units must be standardized during calculations, converting the electric field strength unit from kV / cm to V / m. Then, the formula is used... Calculate the electron drift time.

[0138] Finally, the discharge charge obtained by inversion from S1033 is combined with Using the formula The peak equivalent current of the discharge channel is calculated. Considering that partial discharge is a discrete pulse behavior, and the steady-state temperature rise depends on the average energy per unit time, this embodiment further obtains the discharge pulse repetition frequency statistically analyzed by the UAV within the sampling period. This is obtained from the time resolution and pulse counting module of the multispectral imaging device. The average heat power density correction factor is calculated using this frequency to obtain the average heat power density. .

[0139] S1044. Calculate the ratio of the equivalent current to the cross-sectional area to obtain the scalar current density of the discharge channel, and set the direction of the scalar current density to be consistent with the direction of the electric field vector to obtain the discharge current density.

[0140] Scalar current density refers to the current intensity per unit cross-sectional area, reflecting the concentration of current within the discharge channel. Discharge current density is a vector field, encompassing both magnitude (scalar density) and direction (electric field direction), and is a direct physical quantity for calculating Joule heat power loss.

[0141] In practice, the formula is first used. Calculate the scalar current density. Then, assign this scalar value to each spatial node within the active discharge region, specifying its direction relative to the electric field vector direction at that node. The same principle applies, thus constructing a complete discharge current density vector field. .

[0142] Optionally, the process of determining the density distribution data of discharge heat power loss of the electrical equipment insulation layer based on the discharge current density and the local electric field strength in step S104 may specifically include:

[0143] S1045. By performing a dot product operation on the discharge current density and the local electric field strength, the thermal power density corresponding to each spatial location is obtained.

[0144] Thermal power density refers to the rate density of conversion of electric field energy into heat energy by the directional movement of charges driven by the electric field force and their continuous collisions with gas molecules or dielectric lattices within an active discharge region. Its physical unit is typically 1. .

[0145] In practice, the first step is to target each mesh node in the three-dimensional geometric model that belongs to the active discharge region. Obtain the local electric field intensity vector calculated by S1027. The discharge current density vector calculated by S1044 Next, the instantaneous heat power density of the node is calculated using the vector dot product formula. As shown in the following formula (3):

[0146] (3)

[0147] Since the current direction and the electric field direction are already set to be the same in S1044, the included angle is... The formula simplifies to a product of modulo terms. Finally, the calculated values ​​of all nodes are... The values ​​are stored as a dataset of thermal power density distribution. .

[0148] S1046. Calculate the product of the square of the modulus of the local electric field intensity and the dielectric loss factor of the insulating layer to obtain the loss density generated by the local electric field intensity under the action of the alternating electric field.

[0149] Loss density refers to the energy loss density within an insulating dielectric under alternating current due to repeated polarization, reversing friction, and leakage current of electric dipoles. It is also known as dielectric loss and is widespread throughout the entire insulating volume. The dielectric loss tangent is also mentioned. That is, the loss factor referred to in this application. It is a dimensionless physical quantity that represents the energy loss characteristics of insulating materials, and it depends on the type of material, temperature, and degree of aging. An alternating electric field refers to a sinusoidal electric field that changes periodically with time.

[0150] In practice, the first step is to determine the power grid operating angular frequency based on the equipment register. ,in The operating frequency of the power grid, typically the industrial frequency (50Hz or 60Hz), determines the rate of periodic change of the alternating electric field. It also determines the relative permittivity of the insulating material. and dielectric loss factor Next, traverse each node within the insulation region. Extract its local electric field intensity modulus Finally, the loss density of the node is calculated using the dielectric loss formula. As shown in the following formula (4):

[0151] (4)

[0152] in, Angular frequency, The vacuum permittivity, is the relative permittivity of the material.

[0153] S1047. The thermal power density and loss density are linearly superimposed and mapped to the corresponding spatial nodes in the three-dimensional geometric model to obtain the density distribution data.

[0154] Linear superposition refers to the spatial numerical summation of the two heat sources generated by the aforementioned different mechanisms—namely, the locally intense Joule heating from discharge and the globally distributed dielectric polarization heat—to construct a complete total heat source model. Density distribution data refers to scalar field data that includes the total heat generation rate of all nodes in the three-dimensional geometric model.

[0155] In practice, firstly, an array of total heat sources is initialized, corresponding one-to-one with the mesh nodes of the three-dimensional geometric model. Next, for each node... Determine its location attributes: if the node is located in the active discharge region, then the total heat source... The influence of the duty cycle of the discharge pulse should be considered, i.e. If the node is located in a non-discharge insulating region, then the heat source... .

[0156] Finally, all nodes The values ​​are mapped and bound according to the grid index to generate the final density distribution data of discharge thermal power loss. This embodiment effectively overcomes the limitations of traditional single heat source models and significantly improves the reliability of predicting hidden overheating faults.

[0157] S105. Based on density distribution data and the thermal conductivity and heat dissipation parameters of electrical equipment, the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation is calculated by solving the steady-state heat conduction equation. The steady-state temperature of different areas of the electrical equipment is determined. When there is an area where the steady-state temperature is greater than the preset heat resistance threshold, an early warning message is generated.

[0158] Thermal conductivity and heat dissipation parameters refer to the set of physical parameters describing the thermal characteristics of electrical equipment, including the specific heat capacity, thermal conductivity, and surface convective heat transfer coefficient of materials. The steady-state heat conduction equation is a partial differential equation describing the heat balance relationship when the internal temperature distribution of an object does not change over time; it is usually in the form of the Poisson equation. Temperature field distribution refers to the set of temperature values ​​at all spatial nodes in a three-dimensional geometric model when thermal equilibrium is reached, reflecting the overall heat and cold distribution of the equipment.

[0159] Steady-state temperature refers to the temperature value that eventually stabilizes in each area of ​​the equipment under long-term operating conditions. Preset heat resistance threshold refers to the highest temperature limit at which the insulation material can operate safely for a long period without significant aging or thermal breakdown. Warning information refers to alarm signals issued to maintenance personnel when a potential overheating risk is detected, including the location and severity of the fault, and recommended actions.

[0160] The preset heat resistance threshold is automatically obtained through the system's built-in equipment database: while collecting data, the model information of the target equipment is obtained through RFID tags or image recognition technology, and the corresponding insulation material type is automatically retrieved as shown in Table 3 below. The threshold is then adjusted for aging based on the historical inspection reports of the equipment. For example, for epoxy resin bushings that have been in operation for more than 10 years, the temperature resistance threshold is automatically lowered by 10%, ensuring the personalization and scientific nature of the early warning.

[0161] The preset heat resistance threshold is set according to different types of insulation materials, as shown in Table 3 below:

[0162] Table 3: Comparison Table of Heat Resistance Thresholds

[0163]

[0164] As shown in Table 3, Table 3 lists the temperature resistance grades of commonly used electrical materials and their corresponding critical temperatures, which are used as a benchmark for judging overheating faults.

[0165] In specific implementation, the density distribution data of discharge heat power loss obtained in S1047 is first used to... The insulating region is loaded as a volumetric heat source term into the three-dimensional geometric model. Simultaneously, based on the equipment inventory, the properties of its insulating material, such as epoxy resin, and surface heat dissipation conditions, such as natural convection, are determined, and a steady-state heat conduction equation is constructed, including both internal heat sources and boundary heat dissipation constraints.

[0166] Next, the equation was iteratively calculated using a finite element solver until the temperature at each node of the model no longer changed significantly, thus obtaining the temperature distribution data for the entire field. Subsequently, several key monitoring areas were defined, and the maximum node temperature falling within each area was extracted as the representative steady-state temperature of the corresponding area. And compare it with the preset heat resistance threshold obtained from Table 3. Compare them.

[0167] For example, for epoxy resin sleeves, suppose that the presence of The system immediately marks the area as a potential overheating point and generates an early warning message including the point's three-dimensional coordinates, predicted temperature rise, and suggested power outage maintenance time, which is then sent to the ground monitoring terminal via wireless network.

[0168] This embodiment not only quantifies the hidden temperature rise inside the insulating medium, but also sets a scientific early warning benchmark based on the material's thermal stability characteristics, effectively avoiding the risk of missed detection due to the lag in surface infrared thermography.

[0169] Optionally, the thermal conductivity and heat dissipation parameters include the material's specific heat capacity, thermal conductivity, and heat dissipation coefficient.

[0170] Step S105, which involves calculating the temperature field distribution where internal heat generation and surface heat dissipation reach thermal equilibrium based on density distribution data and the thermal conductivity and dissipation parameters of the electrical equipment, and determining the steady-state temperature of different regions of the electrical equipment by solving the steady-state heat conduction equation, can specifically include:

[0171] Figure 3 A flowchart illustrating a method for determining steady-state temperature according to an embodiment of this application is shown. Figure 3 As shown, the method includes:

[0172] S1051. The density distribution data is loaded as a heat source term into the insulating region of the three-dimensional geometric model, and the specific heat capacity and thermal conductivity of the material are assigned to the insulating region to obtain the insulating entity data.

[0173] The heat source term refers to the volumetric heat source function that generates internal energy in the heat conduction equation. Its value is derived from the density distribution data of the discharge heat power loss generated by S1047. The insulating region refers to the physical domain representing the insulating medium as defined in S1024 above. .

[0174] Material specific heat capacity Thermal conductivity is a physical parameter describing the amount of heat required to raise the temperature of a unit mass of a substance by one unit, usually expressed in J / (kg·K). It is a physical quantity that represents the heat conduction capacity of a material, usually expressed in W / (m·K). Insulation entity data is a comprehensive dataset that includes three-dimensional geometry, volumetric heat source distribution, and material thermal property parameters.

[0175] In practice, the density distribution data obtained from S1047 is first imported. Next, iterate through each mesh node of the insulating region in the 3D geometric model. The total heat source value corresponding to this node Assigned to the heat generation rate parameter in the finite element model At the same time, based on the equipment ledger, such as model B epoxy resin sleeve, its specific heat capacity was obtained by consulting the material manual. and thermal conductivity and give the entire domain To construct virtual entity models, i.e., insulating entity data, that have internal heat sources and thermal conductivity. .

[0176] S1052. Apply the heat dissipation coefficient as a heat transfer boundary condition to the physical interface of the three-dimensional geometric model to obtain boundary condition data.

[0177] Heat dissipation coefficient The term "surface convective heat transfer coefficient" typically refers to the rate at which heat is carried away by a fluid, i.e., air, flowing over a solid surface, and is measured in W / (m²·K). Heat transfer boundary conditions refer to the third type of boundary conditions defined on specific boundaries of the finite element model, used to simulate the heat exchange process between the outer surface of the equipment and its surrounding environment. The physical interface refers to the outer surface of the equipment as defined in S1023 above. Boundary condition data includes complete constraint information such as ambient temperature, heat dissipation coefficient, and the surface on which the action occurs.

[0178] In practice, the current wind speed recorded by S101 is used as the first reference. and ambient temperature The convective heat transfer coefficient of the surface is calculated based on the principle of fluid similarity. The specific steps include: First, based on the characteristic dimensions of the equipment... and air kinematic viscosity Calculate the Reynolds number Subsequently, forced convection correlation was used. Calculate the Nusel number ,in For Prandtl numbers, These are constants related to the shape of the device. Ultimately, through... Surface convective heat transfer coefficient ,in The thermal conductivity of air. The characteristic dimension of the electrical equipment is given, and the heat transfer coefficient is inversely proportional to the characteristic dimension, which is consistent with the fluid boundary layer theory.

[0179] Next, select the physical interface in the finite element analysis software. Apply convective heat flux boundary conditions: ,in, For surface convection heat flux, Let be the surface temperature to be solved. Finally, package the above parameters and boundary definitions to generate boundary condition data. .

[0180] S1053. Based on the data of the insulating entity and the boundary conditions, a steady-state heat conduction equation is constructed according to the law of conservation of energy. The temperature value of each spatial node in the three-dimensional geometric model is obtained by iteratively solving the steady-state heat conduction equation.

[0181] The law of conservation of energy is manifested here as follows: for any infinitesimal element, the heat generated inside it must be equal to the sum of the heat conducted out and the heat lost to the environment, that is, there is no internal energy accumulation under steady state.

[0182] In practice, the aforementioned insulation physical data should be considered first. and boundary condition data Construct a three-dimensional steady-state temperature field The governing equations are shown in the following formula (5):

[0183] (5)

[0184] in, For Hamiltonian operators, Thermal conductivity, For volumetric heat source density, The temperature is to be determined. The equation has boundaries... The aforementioned convective heat transfer conditions must be met. Next, the discretized equations are submitted to the thermal analysis solver, with the initial temperature field set to the ambient temperature. Nonlinear iterative calculations are performed until the maximum temperature residual is less than a preset threshold, such as 0.1℃. Finally, the results for each grid node are output. steady-state temperature value .

[0185] S1054. Map all temperature values ​​to the corresponding spatial locations in the three-dimensional geometric model to obtain the temperature field distribution, and extract the local temperature sets corresponding to different areas of the electrical equipment from the temperature field distribution as the steady-state temperature.

[0186] Local temperature set refers to the temperature subset of all nodes in a specific functional area of ​​the equipment, such as the high-voltage end hardware connection, the root of the shed skirt, and the middle insulator.

[0187] In specific implementation, firstly, the post-processing module is used to process all the values ​​calculated by S1053. Values ​​are mapped back to 3D mesh nodes to generate full-field temperature distribution data. Next, based on the structural characteristics of the equipment, several key monitoring areas are defined, such as the high-voltage end area. Central region and low-voltage end region .

[0188] The node temperatures falling within each region are extracted to form corresponding local temperature sets, such as the high-pressure end temperature set. ,in Belongs to the region The spatial nodes. Similarly, the central temperature set is obtained. and low-pressure end temperature collection Finally, the maximum value in each set is calculated as the representative steady-state temperature of the region. For example, the steady-state temperature at the high-voltage end .

[0189] This embodiment not only quantifies the hidden temperature rise inside the insulating medium, but also sets a scientific early warning benchmark based on the thermal stability characteristics of the material, realizing graded monitoring of different functional areas and effectively avoiding the risk of missed reports due to the lag of surface infrared thermometry.

[0190] Figure 4 This application provides a schematic diagram of a specific implementation of an electrical equipment overheating hazard early warning system based on UAV multispectral imaging, referring to... Figure 4 The system may include:

[0191] The acquisition module 410 is used to acquire photon density data and spatial point cloud data in the ultraviolet band of the surface of electrical equipment collected by a drone equipped with a multispectral imaging device.

[0192] The generation module 420 is used to construct a three-dimensional geometric model based on spatial point cloud data to determine the physical interface of the electrical equipment, and to obtain electric field distribution information including local electric field intensity, electric field vector direction and equipotential surface information by performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment.

[0193] The generation module 420 is also used to obtain the discharge charge and corresponding spatial distribution data of the discharge channel by mapping and correcting the data based on photon density data and electric field distribution information through spatial inversion of local discharge charge.

[0194] The calculation module 430 is used to calculate the discharge current density in the discharge channel based on the discharge charge, spatial distribution data and electric field distribution information at the corresponding location, using the constructed discharge energy dissipation model, and to determine the density distribution data of the discharge heat power loss of the electrical equipment insulation layer based on the discharge current density and local electric field strength.

[0195] The solver module 440 is used to calculate the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation by solving the steady-state heat conduction equation based on density distribution data and the thermal conductivity and heat dissipation parameters of electrical equipment. It determines the steady-state temperature of different areas of electrical equipment and generates early warning information when there is an area where the steady-state temperature is greater than the preset heat resistance threshold.

[0196] The electrical equipment overheating hazard warning system based on UAV multispectral imaging in this application embodiment is used to implement the aforementioned electrical equipment overheating hazard warning method based on UAV multispectral imaging. Therefore, the specific implementation of the electrical equipment overheating hazard warning system based on UAV multispectral imaging can be found in the embodiment section of the electrical equipment overheating hazard warning method based on UAV multispectral imaging mentioned above. The specific implementation can be referred to the description of the corresponding embodiments, and will not be repeated here.

[0197] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application is shown.

[0198] The electronic device may include a processor 510 and a memory 520 storing computer program instructions.

[0199] Specifically, the processor 510 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0200] Memory 520 may include mass storage for data or instructions. For example, and not limitingly, memory 520 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 520 may include removable or non-removable (or fixed) media. Where appropriate, memory 520 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 520 is non-volatile solid-state memory.

[0201] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this disclosure.

[0202] The processor 510 reads and executes computer program instructions stored in the memory 520 to implement any of the above embodiments of the electrical equipment overheating hazard early warning method based on UAV multispectral imaging.

[0203] In one example, the electronic device may also include a communication interface 530 and a bus 540. Wherein, such as Figure 5 As shown, the processor 510, memory 520, and communication interface 530 are connected through bus 540 and complete communication with each other.

[0204] The communication interface 530 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0205] Bus 540 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 540 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0206] The electronic device can execute the electrical equipment overheating hazard warning method based on UAV multispectral imaging in the embodiments of this application, thereby realizing the electrical equipment overheating hazard warning method based on UAV multispectral imaging described in conjunction with the accompanying drawings.

[0207] Furthermore, in conjunction with the above embodiments of the electrical equipment overheating hazard warning method based on UAV multispectral imaging, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the above embodiments of the electrical equipment overheating hazard warning method based on UAV multispectral imaging.

[0208] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0209] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0210] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0211] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0212] The above provides a detailed description of the method and system for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for early warning of overheating hazards in electrical equipment based on UAV multispectral imaging, characterized in that, include: Acquire photon density data and spatial point cloud data in the ultraviolet band from the surface of electrical equipment using a drone equipped with a multispectral imaging device; Based on the spatial point cloud data, a three-dimensional geometric model is constructed to determine the physical interface of the electrical equipment. By performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment, electric field distribution information including local electric field intensity, electric field vector direction and equipotential surface information is obtained. Based on the photon density data and the electric field distribution information, the discharge charge of the discharge channel and the corresponding spatial distribution data are obtained by mapping and correction using the spatial inversion of the local discharge charge. Based on the discharge charge, the spatial distribution data, and the electric field distribution information at the corresponding location, the discharge current density in the discharge channel is calculated using the constructed discharge energy dissipation model. Based on the discharge current density and the local electric field strength, the density distribution data of the discharge heat power loss of the electrical equipment insulation layer is determined. Based on the density distribution data and the thermal conductivity and heat dissipation parameters of the electrical equipment, the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation is calculated by solving the steady-state heat conduction equation. The steady-state temperature of different regions of the electrical equipment is determined, and when there is a region where the steady-state temperature is greater than a preset heat resistance threshold, an early warning message is generated.

2. The method according to claim 1, characterized in that, The step of constructing a three-dimensional geometric model based on the spatial point cloud data to determine the physical interfaces of the electrical equipment includes: The adjacent coordinate points in the spatial point cloud data are topologically connected according to the spatial coordinate adjacency relationship to construct a triangular mesh surface covering the outer contour of the electrical equipment; By closing the triangular mesh surfaces, a three-dimensional geometric model representing the surface topology of the electrical equipment is generated; Based on the geometric morphological characteristics of the three-dimensional geometric model, a solid region is extracted from the three-dimensional geometric model, and the interface between the solid region and the external space is taken as the physical interface.

3. The method according to claim 2, characterized in that, The electric field distribution information, including local electric field intensity, electric field vector direction, and equipotential surface information, is obtained by performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment. The three-dimensional geometric model is divided into an insulating region and an air region based on the physical interface, and the corresponding dielectric parameters are determined for the insulating region and the air region respectively. Based on the operating voltage level and the dielectric parameters, the potential boundary conditions of the insulation region are determined to construct the potential distribution equation; By iteratively solving the potential distribution equation, the potential value of each spatial node is obtained, and the spatial nodes with the same potential value are used as the equipotential surface information. By performing spatial gradient calculation on the potential values, the electric field vector direction and local electric field intensity corresponding to each spatial node are obtained, and all the local electric field intensities, all the electric field vector directions, and the equipotential surface information are combined into the electric field distribution information.

4. The method according to claim 1, characterized in that, The process of obtaining the discharge charge and corresponding spatial distribution data of the discharge channel based on the photon density data and the electric field distribution information by mapping and correcting using the spatial inversion of the local discharge charge includes: Based on the distribution characteristics of the local electric field intensity and the electric field vector direction on the surface of the three-dimensional geometric model, the surface region that satisfies the gas ionization condition is extracted from the surface of the three-dimensional geometric model, and the three-dimensional coordinate set of the surface region is determined as the spatial distribution data; The photon density data is mapped to the spatial location corresponding to the spatial distribution data. The optical path distance from each spatial location to the UAV is calculated to obtain the corresponding attenuation factor. The attenuation factor is used to correct the mapped photon density data to obtain the total number of photons. The total number of photons is converted into the discharge charge of the discharge channel using a preset photoelectric conversion coefficient, which is determined by the gas discharge luminescence mechanism.

5. The method according to claim 1, characterized in that, The calculation of the discharge current density within the discharge channel based on the discharge charge, the spatial distribution data, and the electric field distribution information at the corresponding location, using the constructed discharge energy dissipation model, includes: The connected regions in the spatial distribution data where the local electric field strength is greater than a preset breakdown threshold are identified as active discharge regions, and the maximum span of the active discharge region is calculated along the direction of the electric field vector to obtain the path length of the discharge channel. Calculate the average value of the projected area of ​​the cross-section of the active discharge region to obtain the cross-sectional area of ​​the discharge channel; The equivalent current of the discharge channel is obtained by calculating the ratio of the discharge charge to the electron drift time corresponding to the path length. The ratio of the equivalent current to the cross-sectional area is calculated to obtain the scalar current density of the discharge channel, and the direction of the scalar current density is set to be consistent with the direction of the electric field vector to obtain the discharge current density.

6. The method according to claim 1, characterized in that, The step of determining the density distribution data of discharge heat power loss of the electrical equipment insulation layer based on the discharge current density and the local electric field strength includes: By performing a dot product operation on the discharge current density and the local electric field intensity, the thermal power density corresponding to each spatial location is obtained. The loss density generated by the local electric field intensity under the action of an alternating electric field is obtained by multiplying the square of the modulus of the local electric field intensity with the dielectric loss factor of the insulating layer. The thermal power density and the loss density are linearly superimposed and mapped to the corresponding spatial nodes in the three-dimensional geometric model to obtain the density distribution data.

7. The method according to claim 3, characterized in that, The thermal conductivity and heat dissipation parameters include the material's specific heat capacity, thermal conductivity, and heat dissipation coefficient. Based on the density distribution data and the thermal conductivity and heat dissipation parameters of the electrical equipment, the temperature field distribution in which internal heat generation and surface heat dissipation reach thermal equilibrium is calculated by solving the steady-state heat conduction equation, thereby determining the steady-state temperature of different regions of the electrical equipment, including: The density distribution data is loaded as a heat source term into the insulating region of the three-dimensional geometric model, and the specific heat capacity and thermal conductivity of the material are assigned to the insulating region to obtain the insulating entity data; The heat dissipation coefficient is applied as a heat transfer boundary condition to the physical interface of the three-dimensional geometric model to obtain boundary condition data. Based on the insulating entity data and the boundary condition data, a steady-state heat conduction equation is constructed according to the law of conservation of energy, and the temperature value of each spatial node in the three-dimensional geometric model is obtained by iteratively solving the steady-state heat conduction equation. All the temperature values ​​are mapped to their corresponding spatial locations in the three-dimensional geometric model to obtain the temperature field distribution, and the local temperature sets corresponding to different regions of the electrical equipment are extracted from the temperature field distribution as the steady-state temperature.

8. An early warning system for overheating hazards in electrical equipment based on UAV multispectral imaging, characterized in that, include: The acquisition module is used to acquire photon density data and spatial point cloud data in the ultraviolet band on the surface of electrical equipment collected by a drone equipped with a multispectral imaging device. The generation module is used to construct a three-dimensional geometric model based on the spatial point cloud data to determine the physical interface of the electrical equipment, and to obtain electric field distribution information including local electric field intensity, electric field vector direction and equipotential surface information by performing finite element analysis on the physical interface and the operating voltage level of the electrical equipment. The generation module is also used to obtain the discharge charge and corresponding spatial distribution data of the discharge channel by mapping and correcting the photon density data and the electric field distribution information through spatial inversion of the local discharge charge. The calculation module is used to calculate the discharge current density in the discharge channel based on the discharge charge, the spatial distribution data, and the electric field distribution information at the corresponding location, using the constructed discharge energy dissipation model, and to determine the density distribution data of the discharge heat power loss of the electrical equipment insulation layer based on the discharge current density and the local electric field strength. The solution module is used to calculate the temperature field distribution that achieves thermal equilibrium between internal heat generation and surface heat dissipation by solving the steady-state heat conduction equation based on the density distribution data and the thermal conductivity and heat dissipation parameters of the electrical equipment, determine the steady-state temperature of different regions of the electrical equipment, and generate early warning information when there is a region where the steady-state temperature is greater than a preset heat resistance threshold.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for early warning of electrical equipment overheating hazards based on UAV multispectral imaging as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of the electrical equipment overheating hazard early warning method based on UAV multispectral imaging as described in any one of claims 1 to 7.

Images