Modeling method and system for power equipment

Abstract

The invention relates to the technical field of electric power, in particular to a modeling method and system for electric power equipment. The method comprises the following steps: acquiring a three-dimensional tomographic image set of a contact and constructing an initial simple three-dimensional model of the contact; mutual information maximization registration is carried out on the contact three-dimensional tomography image set, and a contact registration image data set is obtained; recognizing contact deformation of the high-voltage circuit breaker based on the contact registration image data set, and generating a contact surface deformation distribution diagram; constructing a contact multi-scale DGDI pseudo-color graph based on the contact surface deformation distribution graph; generating a contact multi-scale deformation gradient map based on the contact multi-scale DGDI pseudo-color map; constructing a contact deformation displacement field based on the contact multi-scale deformation gradient map; and constructing a contact layered stress transfer model. According to the invention, the precision and reliability of the construction of the contact motion trail model are obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power, and in particular, to a modeling method and system for electric power equipment. Background Art

[0002] In the field of on-line monitoring of electric power equipment, such as high-voltage circuit breakers, there are significant problems of accuracy loss in the dynamic correction of the contact motion trajectory in the existing modeling methods. When the circuit breaker performs an opening operation, due to the coupling effect of mechanical stress and electromagnetic load during the high-speed movement of the contacts, asymmetric plastic deformation will occur, resulting in a systematic deviation between the actual contact surface geometry and the theoretical model. This deviation is particularly prominent after multiple mechanical operations - due to the cumulative effect of contact material fatigue and micro-defects, the deformation area gradually expands from local to global, making it difficult for the motion model based on ideal geometric assumptions to represent the micro-deformation gradient characteristics during the dynamic separation process of the contacts. The main technical bottlenecks of the existing methods are reflected in three aspects: First, the multi-scale representation of contact deformation lacks systematicness, and the correlation mapping of deformation gradients has not been established at the micro (grain level), meso (defect level), and macro (component level) scales, resulting in insufficient quantization accuracy of deformation characteristics; Second, the modeling of the stress transfer mechanism is too simplified, and the co-evolution law of the material nonlinear constitutive relationship and the layered stress distribution (such as elastic region, plastic region, and interface slip region) has not been considered; Third, the correlation analysis of deformation data and motion parameters (such as displacement, acceleration) only stays at the static matching level, and a dynamic feedback mechanism has not been constructed to correct the motion trajectory model. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a modeling method and system for electric power equipment to solve at least one of the above technical problems.

[0004] To achieve the above object, a modeling method for electric power equipment is applied to the moving contact of a high-voltage circuit breaker. The modeling method for electric power equipment includes the following steps:

[0005] Step S1: Collect a set of three-dimensional tomography images of the contact and construct an initial simple three-dimensional model of the contact; perform mutual information maximization registration on the set of three-dimensional tomography images of the contact to obtain a data set of registered contact images; identify the contact deformation of the high-voltage circuit breaker based on the data set of registered contact images, and generate a contact surface deformation distribution map;

[0006] Step S2: Construct a multi-scale DGDI pseudo-color map of the contact based on the contact surface deformation distribution map;

[0007] Step S3: Generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact; construct a contact deformation displacement field based on the multi-scale deformation gradient map of the contact; construct a contact layered stress transfer model;

[0008] Step S4: Conduct contact material response analysis based on the contact deformation displacement field and the contact delamination stress transfer model to obtain a contact material response characteristic map; conduct contact deformation simulation based on the contact material response characteristic map and the contact delamination stress transfer model to obtain contact deformation co-evolution data;

[0009] Step S5: Perform motion trajectory mapping on the initial simple 3D model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; conduct motion simulation on the initial simple 3D model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

[0010] By collecting a set of 3D tomography images of the contact and performing mutual information maximization registration, the present invention can accurately identify the deformation of the high-voltage circuit breaker contact, effectively solving the problem of systematic deviation between the actual contact surface geometry and the theoretical model caused by contact deformation in the prior art. Secondly, by constructing a multi-scale DGDI pseudo-color map of the contact based on the contact surface deformation distribution map, a multi-scale representation of contact deformation is realized, which can establish a correlation mapping of deformation gradients at the micro (grain level), meso (defect level), and macro (component level) scales, significantly improving the quantization accuracy of deformation characteristics and overcoming the deficiencies of lack of systematicness and insufficient quantization accuracy in multi-scale representation in the prior art. By constructing a contact delamination stress transfer model, the co-evolution law of the material nonlinear constitutive relationship and the delamination stress distribution (such as the elastic zone, plastic zone, and interface slip zone) is considered, making the modeling of the stress transfer mechanism more accurate and compensating for the deficiency of overly simplified modeling of the stress transfer mechanism in the existing methods. By conducting contact material response analysis and contact deformation simulation based on the contact deformation displacement field and the contact delamination stress transfer model, and accordingly performing motion trajectory mapping and motion simulation on the initial simple 3D model of the contact, a dynamic feedback mechanism is constructed, which can real-time correct the motion trajectory model, solving the problem that the correlation analysis between deformation data and motion parameters in the prior art only stays at the static matching level and lacks a dynamic feedback mechanism, thereby improving the accuracy of dynamic correction of the contact motion trajectory.

[0011] Preferably, the collecting a set of 3D tomography images of the contact in Step S1 is specifically:

[0012] Use a laser interferometer to measure the high-voltage circuit breaker contact at multiple positions to obtain a set of geometric parameters of the high-voltage circuit breaker contact;

[0013] Construct an initial simple 3D model of the contact based on the set of geometric parameters of the high-voltage circuit breaker contact;

[0014] Construct a laser scanning path based on the initial simple 3D model of the contact;

[0015] The contact of the high-voltage circuit breaker is scanned layer by layer at multiple angles according to the laser scanning path to obtain a three-dimensional tomography image set of the contact. Among them, the multi-angle layer scanning includes optical coherence tomography imaging at 0°, 45°, 90°, 135°, and 180° azimuths. The three-dimensional tomography image set of the contact includes tomography slice data with a transverse resolution of 3 μm and a longitudinal resolution of 1 μm of the contact surface topography.

[0016] In the present invention, by using a laser interferometer to measure the high-voltage circuit breaker contact at multiple positions and constructing an initial simple three-dimensional model of the contact, it can provide an accurate reference basis for subsequent laser scanning path planning. By adopting multi-angle layer scanning, including optical coherence tomography imaging at 0°, 45°, 90°, 135°, and 180° azimuths, the comprehensiveness and accuracy of the three-dimensional tomography image set of the contact are ensured. The obtained contact surface topography data has high transverse and longitudinal resolutions (transverse resolution 3m, longitudinal resolution 1m), and can effectively capture the micro-deformation characteristics and detailed information on the contact surface.

[0017] Preferably, the mutual information maximization registration of the three-dimensional tomography image set of the contact in step S1 is specifically as follows:

[0018] Calculate the mutual information values between different-angle images in the three-dimensional tomography image set of the contact. Among them, the specific calculation formula is as follows;

[0019]

[0020] Among them, I(A, B) is the mutual information value between the reference image A and the image B to be registered, p(a, b) is the joint probability distribution of the gray values of the two images at the same spatial position, p(a) is the marginal probability distribution of the reference image A, p(b) is the marginal probability distribution of the image B to be registered, a is the gray value of the reference image A, and b is the gray value of the image B to be registered;

[0021] Maximize the mutual information value to obtain the maximized mutual information; among them, the maximization is specifically:

[0022] By adjusting the translation parameters (t′ x , t′ y , t′ z ) and rotation parameters (θ x , θ y , θ z ) of the image B to be registered;

[0023] The parameter update formula for adjusting the image B to be registered is:

[0024]

[0025] Among them, θ is the parameter vector to be optimized, including the translation parameter (t′x , t′ y , t′ z ), and the rotation parameters (θ x , θ y , θ z ), t′ x , t′ y , t′ z are the translation amounts of the image in the X, Y, and Z axis directions respectively, and θ x , θ y , θ z are the rotation angles of the image around the X, Y, and Z axes respectively, and α is the step size for controlling parameter update. is the gradient of the mutual information I(A, B) with respect to the parameter θ;

[0026] Apply θ new to the image B to be registered for spatial transformation to obtain the aligned image B';

[0027] Overlay the reference image A and all the aligned images B' according to their spatial positions to generate a contact registration image dataset.

[0028] By adopting the mutual information maximization registration method, the present invention can efficiently and accurately align the multi-angle images in the contact three-dimensional tomography image set. The calculation of the mutual information value fully considers the joint probability distribution of the image gray values, enabling the registration process to effectively capture the internal correlation between images, thereby achieving the best match between different angle images. By adjusting the translation parameters and rotation parameters and optimizing them using the parameter update formula, the accuracy and efficiency of image registration are further improved. The finally generated contact registration image dataset can accurately reflect the actual geometric shape of the contact.

[0029] Preferably, identifying the contact deformation of the high-voltage circuit breaker based on the contact registration image dataset in step S1 specifically includes:

[0030] Perform multi-modal gray normalization on the contact registration image dataset to obtain a contact multi-modal gray normalization image set;

[0031] Perform contact surface defect segmentation on the contact multi-modal gray normalization image set to obtain a contact surface defect segmentation label map;

[0032] Perform dilation processing on the defect regions in the contact surface defect segmentation label map to obtain a contact surface defect dilation map; perform a closing operation on the defect regions of the contact surface defect dilation map to obtain a contact surface defect closed segmentation map;

[0033] Extract the contact spiral chain code sequence based on the contact surface defect closed segmentation map to generate a contact surface defect spiral chain code set;

[0034] Perform dynamic expansion of the defect neighborhood based on the spiral chain code set of the contact surface defects and the laser scanning path to generate an extended mask image of the contact surface defects;

[0035] Calculate the directional gradient field based on the multi-modal grayscale normalized image set of the contact and the extended mask image of the contact surface defects to obtain the directional gradient distribution map of the contact surface;

[0036] Calculate the deformation curvature of the contact surface based on the directional gradient distribution map of the contact surface, and generate the curvature distribution map of the contact surface according to the deformation curvature of the contact surface;

[0037] Generate the deformation intensity distribution data of the contact surface based on the curvature distribution map of the contact surface and the directional gradient distribution map of the contact surface;

[0038] Perform weighted fusion of the deformation intensity distribution data of the contact surface and the extended mask image of the contact surface defects to generate the deformation distribution map of the contact surface;

[0039] Perform non-maximum suppression on the deformation distribution map of the contact surface to obtain the deformation feature map of the contact surface;

[0040] Perform skeleton extraction on the deformation feature map of the contact surface to generate the deformation distribution map of the contact surface.

[0041] Through multi-modal grayscale normalization processing, the present invention can eliminate the grayscale differences of images under different imaging conditions and provide a unified image basis for subsequent analysis. Through the segmentation and closing operation of the contact surface defects, the defect area can be accurately extracted and the existing segmentation gaps can be filled to ensure the integrity of the defect area. Through the extraction of the spiral chain code sequence and the dynamic expansion of the defect neighborhood, the boundary and scope of the defect are further clarified, providing accurate regional positioning for subsequent deformation analysis. Through the generation of the directional gradient field and the curvature distribution map, the deformation characteristics of the contact surface can be reflected from different angles, including height changes and curvature changes, so as to capture deformation information more comprehensively. Through the weighted fusion of the deformation intensity distribution data and the defect extended mask image, the deformation characteristics of the defect area can be highlighted while retaining the overall deformation information, making the deformation distribution more intuitive. Through non-maximum suppression and skeleton extraction, the deformation characteristics are further simplified, and the most significant deformation characteristics are extracted for subsequent analysis and modeling.

[0042] Preferably, step S2 includes the following steps:

[0043] Step S21: Perform Gaussian blur on the deformation distribution map of the contact surface according to a preset scale to generate the geometric deformation data of the contact surface at the micro scale, the geometric deformation data of the contact surface at the meso scale, and the geometric deformation data of the contact surface at the macro scale, where the preset scale includes the micro scale, the meso scale, and the macro scale;

[0044] Step S22: Calculate the mean and standard deviation of the local contact surface geometric deformation index at each scale;

[0045] Step S23: Map the contact micro-scale geometric deformation data, contact meso-scale geometric deformation data, and contact macro-scale geometric deformation data to the RGB three channels respectively according to the mean and standard deviation, and perform fusion to generate a contact multi-scale DGDI pseudo-color map.

[0046] The present invention generates deformation data at different scales through Gaussian blur processing, which can effectively capture the characteristic differences of contact deformation at the micro, meso, and macro levels, and avoid information loss caused by single-scale analysis. Calculating the mean and standard deviation of the geometric deformation index at each scale further quantifies the distribution law and change trend of the deformation characteristics. By mapping the deformation data at different scales to the RGB three channels and fusing them to generate a pseudo-color map, not only the visual identifiability of the deformation characteristics is enhanced, but also the intensity and distribution details of the deformation are highlighted through the adjustment of brightness and transparency, making the complex deformation information more intuitive and understandable. This provides a richer and more accurate basis for the health state assessment of high-voltage circuit breaker contacts.

[0047] Preferably, in step S3, generating a contact multi-scale deformation gradient map based on the contact multi-scale DGDI pseudo-color map specifically includes:

[0048] Obtain the contact surface curvature distribution map;

[0049] Identify the contact deformation gradient direction angle based on the contact surface curvature distribution map, generate a vector arrow through the contact deformation gradient direction angle and the contact surface geometric deformation index, where the length represents the deformation intensity and the direction represents the deformation trend, to generate a contact surface deformation gradient vector field. The specific calculation formula of the contact deformation gradient direction angle is as follows:

[0050]

[0051] where θ(x,y) is the deformation gradient direction angle, is the first-order partial derivative of the surface height function in the x direction, is the first-order partial derivative of the surface height function in the y direction;

[0052] Perform spatial vector superposition on the contact surface deformation distribution map, the contact multi-scale DGDI pseudo-color map, and the contact surface deformation gradient vector field based on a preset unified spatial coordinate system to generate a contact multi-scale deformation gradient map.

[0053] By identifying the contact deformation gradient direction angle and generating a deformation gradient vector field, the present invention can intuitively display the intensity and trend of deformation, providing a powerful tool for analyzing the propagation direction and evolution law of deformation. By performing spatial vector superposition on the contact surface deformation distribution map, multi-scale DGDI pseudo-color map and deformation gradient vector field, the generated multi-scale deformation gradient map not only retains the multi-scale detailed information of deformation, but also clearly presents the dynamic characteristics of deformation in the form of a vector field. This can more accurately reflect the deformation behavior of the contact under complex working conditions, providing richer and more accurate data support for the health monitoring and fault prediction of high-voltage circuit breaker contacts.

[0054] Preferably, in step S3, constructing the contact deformation displacement field based on the multi-scale deformation gradient map of the contact is specifically as follows:

[0055] Extract the sub-pixel level texture features in the multi-scale deformation gradient map of the contact to generate a local binary pattern coding map of the contact;

[0056] Perform mapping fusion on the multi-scale deformation gradient map of the contact and the local binary pattern coding map of the contact to obtain a microscopic deformation feature map of the contact;

[0057] Obtain the initial morphology reference map of the high-voltage circuit breaker contact;

[0058] Based on the initial morphology reference map of the high-voltage circuit breaker contact, identify the deformation feature points in the microscopic deformation feature map of the contact to obtain a set of contact deformation feature difference coordinates;

[0059] Based on the initial morphology reference map of the high-voltage circuit breaker contact and the microscopic deformation feature map of the contact, perform displacement field inversion on the set of contact deformation feature difference coordinates to obtain the contact deformation displacement field.

[0060] By extracting the sub-pixel level texture features in the multi-scale deformation gradient map and generating a local binary pattern coding map, the present invention can capture the minute deformation details on the contact surface, thereby achieving high-precision identification of deformation features. By performing mapping fusion on the deformation gradient map and the local binary pattern coding map, the visualization effect of deformation features is further enhanced, making the microscopic deformation features more obvious. Based on the initial morphology reference map, identifying deformation feature points and performing displacement field inversion can accurately determine the position and degree of deformation, and the generated deformation displacement field distribution map intuitively reflects the overall situation of contact deformation. This can not only effectively identify the microscopic deformation of the contact, but also provide accurate data support for the dynamic monitoring and health assessment of the contact.

[0061] By collecting the measured stress-strain curve and optimizing the parameters of the Ramberg-Osgood constitutive model, the present invention ensures the accurate fitting of the model to the nonlinear characteristics of the contact material, thereby providing a reliable theoretical basis for stress transfer analysis. Further, stress transfer models are constructed for the micro, meso, and macro scales respectively, which can not only reflect the deformation characteristics of the material at different scales, but also quantify the slip effect of the grain boundary or defect region through the interface slip correction coefficient, enabling the model to more realistically reflect the stress distribution and deformation behavior of the contact under actual working conditions. This provides a more scientific and accurate mechanical basis for the deformation simulation and health monitoring of the contact, and helps to deeply understand the failure mechanism of the contact.

[0062] Preferably, step S4 includes the following steps:

[0063] Step S41: Inverting the contact surface strain tensor through finite element numerical simulation based on the contact deformation displacement field and the contact hierarchical stress transfer model;

[0064] Step S42: Solving the contact surface strain tensor through the contact hierarchical stress transfer model to obtain the contact multi-scale stress field data;

[0065] Step S43: Fusing the contact surface strain tensor and the contact multi-scale stress field data to obtain the contact material response characteristic map;

[0066] Step S44: Conducting contact motion simulation on the high-voltage circuit breaker contact based on the contact material response characteristic map and the contact hierarchical stress transfer model to obtain the contact dynamic stress field data;

[0067] Step S45: Constructing a contact nonlinear deformation response model based on the contact dynamic stress field data;

[0068] Step S46: Collecting the actual working condition parameters of the contact; conducting contact deformation simulation on the contact nonlinear deformation response model according to the actual working condition parameters of the contact to obtain the contact deformation co-evolution data.

[0069] By inverting the contact surface strain tensor through finite element numerical simulation and combining with the hierarchical stress transfer model, the present invention can accurately solve the stress field distribution of the contact at multiple scales, providing key data support for deeply understanding the mechanical response of the contact material. Further, fusing the strain tensor and the multi-scale stress field data to generate the material response characteristic map provides a rich information basis for the dynamic behavior analysis of the contact. Based on this, obtaining the dynamic stress field data through contact motion simulation and constructing a nonlinear deformation response model can accurately describe the dynamic deformation process of the contact under complex working conditions. Finally, combining the actual working condition parameters for deformation simulation, the obtained deformation co-evolution data can provide a scientific basis for the health monitoring and life prediction of the contact, significantly improving the prediction accuracy and reliability of the contact deformation behavior.

[0070] Preferably, step S5 includes the following steps:

[0071] Step S51: Detect and process outliers in the contact deformation co-evolution data to obtain standard deformation evolution data;

[0072] Step S52: Extract feature vectors from the standard deformation evolution data to obtain the contact motion trajectory feature vectors;

[0073] Step S53: Obtain the initial simple 3D model of the contact; perform coordinate system transformation and parameter mapping on the contact motion trajectory feature vectors based on the initial simple 3D model of the contact to obtain the contact motion parameter mapping space;

[0074] Step S54: Perform correlation modeling on the contact motion parameter mapping space to obtain the contact motion parameter correlation data;

[0075] Step S55: Screen the key motion parameters of the contact motion trajectory feature vectors according to the contact motion parameter correlation data to obtain the contact motion parameter candidate set;

[0076] Step S56: Import the contact motion parameter candidate set into the initial simple 3D model of the contact and perform motion trajectory modeling to obtain the contact motion trajectory model.

[0077] By detecting outliers and extracting feature vectors from the deformation co-evolution data, the present invention can effectively remove noise data and extract key features, ensuring the accuracy of subsequent modeling. Coordinate system transformation and parameter mapping of the feature vectors based on the initial 3D model generate a motion parameter mapping space that provides a unified reference framework for subsequent analysis. Through correlation modeling and key parameter screening, the selection of motion parameters is further optimized, redundant information is removed, and the efficiency and reliability of the model are improved. Finally, based on the optimized parameter set, motion trajectory modeling is performed, and the generated contact motion trajectory model can more realistically reflect the dynamic behavior of the contact under actual working conditions, providing important technical support for the health monitoring, fault prediction, and maintenance strategy optimization of power equipment.

[0078] Preferably, the present invention also provides a modeling system for power equipment, which is used to execute the modeling method for power equipment as described above. The modeling system for power equipment includes:

[0079] An image registration module, which is used to collect a set of contact three-dimensional tomography images and construct an initial simple 3D model of the contact; perform mutual information maximization registration on the set of contact three-dimensional tomography images to obtain a contact registration image data set; identify the contact deformation of the high-voltage circuit breaker based on the contact registration image data set and generate a contact surface deformation distribution map;

[0080] The deformation analysis module is used to construct a multi-scale DGDI pseudo-color map of the contact based on the deformation distribution map of the contact surface;

[0081] The model construction module is used to generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact; construct a deformation displacement field of the contact based on the multi-scale deformation gradient map of the contact; construct a hierarchical stress transfer model of the contact;

[0082] The response simulation module is used to perform contact material response analysis based on the contact deformation displacement field and the contact hierarchical stress transfer model to obtain a contact material response characteristic map; perform contact deformation simulation based on the contact material response characteristic map and the contact hierarchical stress transfer model to obtain contact deformation co-evolution data;

[0083] The contact motion modeling module is used to perform motion trajectory mapping on the initial simple 3D model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; perform motion simulation on the initial simple 3D model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

[0084] The present invention efficiently completes the preprocessing and registration of image data through the image registration module to ensure the accuracy of subsequent analysis; the deformation analysis module and the model construction module provide detailed deformation characteristic data through multi-scale analysis and modeling, and at the same time generate a multi-scale DGDI pseudo-color map and a deformation gradient map, accurately capturing the deformation characteristics of the contact at different scales, significantly improving the accuracy and reliability of deformation monitoring, and helping to detect potential problems early and reduce the risk of equipment failure. The response simulation module and the contact motion modeling module further generate a contact motion trajectory model through dynamic simulation and motion analysis, and realize real-time correction and dynamic feedback of the contact motion trajectory, more accurately reflecting the dynamic behavior of the contact under actual working conditions. Description of the Drawings

[0085] Other features, objects, and advantages of the present invention will become more apparent by reading the following detailed description with reference to the accompanying drawings:

[0086] Figure 1 The step flow diagram of the modeling method for power equipment according to an embodiment is shown.

[0087] Figure 2 The detailed step flow diagram of step S2 according to an embodiment is shown.

[0088] Figure 3 The detailed step flow diagram of step S4 according to an embodiment is shown. Detailed Embodiment

[0089] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0090] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0091] It should be understood that although terms such as "first" and "second" may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit can be called the second unit, and similarly the second unit can be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed associated items.

[0092] To achieve the above object, please refer to Figures 1 to 3 , the present invention provides a modeling method for power equipment, which is applied to the moving contact of a high-voltage circuit breaker. The modeling method for power equipment includes the following steps:

[0093] Step S1: Collect a set of three-dimensional tomography images of the contact and construct an initial simple three-dimensional model of the contact; perform mutual information maximization registration on the set of three-dimensional tomography images of the contact to obtain a dataset of registered contact images; identify the deformation of the contact of the high-voltage circuit breaker based on the dataset of registered contact images and generate a distribution map of the surface deformation of the contact.

[0094] Step S2: Construct a multi-scale DGDI pseudocolor map of the contact based on the distribution map of the surface deformation of the contact.

[0095] Step S3: Generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudocolor map of the contact; construct a deformation displacement field of the contact based on the multi-scale deformation gradient map of the contact; construct a hierarchical stress transfer model of the contact.

[0096] Step S4: Perform contact material response analysis based on the contact deformation displacement field and the contact delamination stress transfer model to obtain a contact material response characteristic map; perform contact deformation simulation based on the contact material response characteristic map and the contact delamination stress transfer model to obtain contact deformation co-evolution data;

[0097] Step S5: Perform motion trajectory mapping on the initial simple 3D model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; perform motion simulation on the initial simple 3D model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

[0098] In this embodiment, a Konica Minolta Vivid 910 laser scanner is used to perform multi-angle delamination scanning on the contacts of an SF6 high-voltage circuit breaker to collect a set of 3D tomographic images of the contacts. This image set contains detailed information on the surface topography of the contacts, with a lateral resolution of 3 microns and a longitudinal resolution of 1 micron. These images are imported into MATLAB software, and the mutual information maximization registration algorithm is used to process the image set. By calculating the mutual information values between images at different angles and adjusting the translation and rotation parameters of the images to maximize the mutual information value, a contact registration image data set is generated. Next, Adobe Photoshop is used to perform contact surface defect segmentation on the registered images to generate a defect segmentation marker map, and MATLAB is used to extend the defect boundary to a surrounding area of 5 microns to form a defect extension mask map. Based on this mask map and the registration image data set, the local curvature of the contact defect area is calculated to generate a contact surface curvature distribution map. The Sobel operator is used to perform convolution operation on the curvature distribution map to generate a contact surface curvature gradient map, and based on this, the contact surface geometric deformation index is calculated to construct a contact multi-scale DGDI pseudo-color map. Further, using the image processing function of MATLAB, a contact multi-scale deformation gradient map is generated based on the DGDI pseudo-color map, and by analyzing the deformation gradient direction angle and the geometric deformation index, the deformation of the contact is identified to generate a contact deformation displacement field. On this basis, the measured stress-strain curve of the contact material is collected, and a Ramberg-Osgood constitutive model is constructed in MATLAB. The model parameters are optimized by the least squares method to construct a contact delamination stress transfer model. Using this model and the deformation displacement field distribution map, contact material response analysis is performed in ANSYS finite element analysis software to obtain a contact material response characteristic map. Based on this characteristic map and the delamination stress transfer model, contact motion simulation is performed in ANSYS to obtain contact deformation co-evolution data. Finally, the initial simple 3D model of the contact is obtained, and in SolidWorks, motion trajectory mapping is performed on the model according to the deformation co-evolution data to obtain a set of contact motion parameters. Using these parameter sets, motion simulation is performed on the initial simple 3D model of the contact in SolidWorks Motion to generate a contact motion trajectory model.

[0099] Preferably, the acquisition of the three-dimensional tomographic image set of the contact in step S1 is specifically as follows:

[0100] Use a laser interferometer to measure the high-voltage circuit breaker contact at multiple positions to obtain a set of geometric parameters of the high-voltage circuit breaker contact;

[0101] Construct an initial simple three-dimensional model of the contact based on the set of geometric parameters of the high-voltage circuit breaker contact;

[0102] Construct a laser scanning path based on the initial simple three-dimensional model of the contact;

[0103] Perform multi-angle layer scanning on the high-voltage circuit breaker contact according to the laser scanning path to obtain a three-dimensional tomographic image set of the contact. Among them, the multi-angle layer scanning includes optical coherence tomography imaging in the azimuths of 0°, 45°, 90°, 135°, and 180°. The three-dimensional tomographic image set of the contact includes tomographic slice data with a lateral resolution of 3 μm and a longitudinal resolution of 1 μm of the contact surface topography.

[0104] Specifically, a laser interferometer (such as Zygo NewView 9300) is used to perform multi-position measurements on the contacts of an SF6 high-voltage circuit breaker. The laser interferometer is fixed on a tripod, and by adjusting its position and angle, it can cover all key parts of the high-voltage circuit breaker contacts. Five measurement positions are selected, which are located at the top, bottom, left, right, and center positions of the contacts respectively. At each position, the laser interferometer emits a laser beam onto the contact surface and receives the reflected optical signal, and uses the interference principle to accurately measure the geometric parameters of the contact surface. After multiple measurements, a set of geometric parameters of the contacts is obtained, including data such as the size, shape, and surface roughness of the contacts. The set of geometric parameters of the contacts obtained from the laser interferometer is imported into 3D modeling software (such as SolidWorks), and according to the size and shape data in the geometric parameter set, a simple initial 3D model of the contacts is manually created. This model is a rough geometric shape, only containing the basic outer shape and main structural features of the contacts, without complex details. Referring to the technical manual and design drawings of the high-voltage circuit breaker, the dimensions of the model are calibrated and adjusted. Finally, the simple initial 3D model of the contacts is saved in a standard 3D file format. The simple initial 3D model of the contacts is imported into laser scanning path planning software (such as Rapidform XOR), and according to the geometric shape and size of the contacts, combined with the technical parameters of the laser scanning device (such as Konica Minolta Vivid 910), a laser scanning path is manually planned. This path covers all key areas of the contact surface, including the contact surface, edges, and connection parts of the contacts. The spiral scanning mode is selected to ensure the uniformity and continuity of the scanning. At the same time, parameters such as the starting point, ending point, and scanning speed of the scanning are set, and the scanning speed is selected as 50 millimeters per second according to the surface characteristics of the contacts. After the planning is completed, the scanning path is exported in a dedicated path file format (such as.GCODE format) for guiding the scanning operation of the laser scanning device. The high-voltage circuit breaker contacts are placed on the workbench of the laser scanning device (Konica Minolta Vivid 910), and multi-angle layer scanning is performed according to the previously planned laser scanning path file. The scanning device performs optical coherence tomography on the contacts from five azimuths of 0°, 45°, 90°, 135°, and 180° respectively according to the preset path. At each azimuth, the scanning device emits a laser beam and receives the reflected optical signal to obtain the tomographic slice data of the contact surface topography. The lateral resolution of the scanning is set to 3 microns, and the longitudinal resolution is set to 1 micron. After the scanning is completed, the tomographic slice data of all azimuths are integrated together to form a 3D tomographic image set of the contacts.

[0105] Preferably, the mutual information maximization registration of the 3D tomographic image set of the contacts in step S1 is specifically as follows:

[0106] Calculate the mutual information value between images at different angles in the three-dimensional tomographic image set of the contact, and the specific calculation formula is as follows;

[0107]

[0108] Among them, I(A, B) is the mutual information value between the reference image A and the image B to be registered, p(a, b) is the joint probability distribution of the gray values of the two images at the same spatial position, p(a) is the marginal probability distribution of the reference image A, p(b) is the marginal probability distribution of the image B to be registered, a is the gray value of the reference image A, and b is the gray value of the image B to be registered;

[0109] Maximize the mutual information value to obtain the maximized mutual information; among them, the maximization is specifically:

[0110] By adjusting the translation parameters (t′ x , t′ y , t′ z ) and rotation parameters (θ x , θ y , θ z ) of the image B to be registered;

[0111] The parameter update formula for adjusting the image B to be registered is:

[0112]

[0113] Among them, θ is the parameter vector to be optimized, including translation parameters (t′ x , t′ y , t′ z ) and rotation parameters (θ x , θ y , θ z ), t′ x , t′ y , t′ z are the translation amounts of the image in the X, Y, and Z axis directions respectively, θ x , θ y , θ z are the rotation angles of the image around the X, Y, and Z axes respectively, α is the step size for controlling parameter update, is the gradient of the mutual information I(A, B) with respect to the parameter θ;

[0114] Apply θ new to the image B to be registered for spatial transformation to obtain the aligned image B';

[0115] Overlay the reference image A and all the aligned images B' according to the spatial position to generate a contact registration image data set.

[0116] Specifically, MATLAB software can be used to process the three-dimensional tomography image sets of the contact obtained from different angles (0°, 45°, 90°, 135°, 180°). First, load these images and write MATLAB scripts to calculate the mutual information value. By using the imhist and histcounts functions in MATLAB, the joint probability distribution p(a,b) of the gray values at the same spatial position of two images, as well as the marginal probability distributions p(a) and p(b) of the reference image A and the image B to be registered, are calculated. According to the formula By looping through all combinations of gray values, the mutual information value is calculated. In this process, a clear 0° image is selected as the reference image A, and the images at other angles are used as the images B to be registered. The mutual information values between them and the reference image are calculated respectively. Then, an optimization objective function is defined, that is, to maximize the mutual information value, and a gradient-based optimization algorithm is adopted. Initialize the translation parameters t′ x ,t′ y ,t′ x and the rotation parameter θ x ,θ y ,θ z to be zero, and then calculate the gradients of these parameters with respect to the mutual information and adjust the parameters step by step according to the parameter update formula . Select the step size α = 0.01, which is a suitable step size value verified through multiple experiments and can ensure the stability and convergence speed of the optimization process. After each parameter update, use the imwarp function in MATLAB to perform a spatial transformation on the image B to be registered to generate the aligned image B'. Through continuous iterative optimization, the mutual information value is finally maximized, achieving precise alignment between images at different angles. After image registration is completed, the reference image A is superimposed with all the aligned images B'. Use the imfuse function in MATLAB to superimpose these images according to the spatial position. To ensure that the superimposed image can clearly display the geometric morphology of the contact, adjust the transparency and contrast parameters of the superimposed image. Finally, a complete contact registration image data set is generated, which contains contact images obtained from different angles and these images are completely aligned in spatial position. Among them, traverse all pixel pairs in the reference image A and the image B to be registered, count the occurrence frequency of the gray value pairs (a,b) to obtain the joint probability distribution p(a,b); separately count the gray histograms of the reference image A and the image B to be registered to obtain p(a) and p(b).

[0117] Preferably, identifying the deformation of the contact of the high-voltage circuit breaker based on the contact registration image data set in step S1 is specifically as follows:

[0118] Perform multi-modal gray normalization on the contact registration image dataset to obtain the contact multi-modal gray normalized image set;

[0119] Perform contact surface defect segmentation on the contact multi-modal gray normalized image set to obtain the contact surface defect segmentation label map;

[0120] Perform dilation on the defect regions in the contact surface defect segmentation label map to obtain the contact surface defect dilation map; perform a closing operation on the defect regions of the contact surface defect dilation map to obtain the contact surface defect closed segmentation map;

[0121] Extract the contact spiral chain code sequence based on the contact surface defect closed segmentation map to generate the contact surface defect spiral chain code set;

[0122] Perform defect neighborhood dynamic expansion according to the contact surface defect spiral chain code set and the laser scanning path to generate the contact surface defect expansion mask map;

[0123] Calculate the directional gradient field based on the contact multi-modal gray normalized image set and the contact surface defect expansion mask map to obtain the contact surface directional gradient distribution map;

[0124] Calculate the contact surface deformation curvature based on the contact surface directional gradient distribution map, and generate the contact surface curvature distribution map according to the contact surface deformation curvature;

[0125] Generate the contact surface deformation intensity distribution data based on the contact surface curvature distribution map and the contact surface directional gradient distribution map;

[0126] Perform weighted fusion of the contact surface deformation intensity distribution data and the contact surface defect expansion mask map to generate the contact surface deformation distribution map;

[0127] Perform non-maximum suppression on the contact surface deformation distribution map to obtain the contact surface deformation feature map;

[0128] Perform skeleton extraction on the contact surface deformation feature map to generate the contact surface deformation distribution map.

[0129] Specifically, MATLAB software can be used to perform multi-modal gray normalization on the contact registration image dataset. First, load the contact registration image dataset. In MATLAB, call the imhist function to analyze the gray histogram of the image and determine the range of gray values. Then use the imadjust function to normalize the image and adjust the gray value range to [0, 255]. For example, for an image with a gray value range of [50, 200], by setting the parameters of the imadjust function, its gray value range is linearly mapped to [0, 255], thus realizing the gray normalization of the multi-modal image and obtaining the contact multi-modal gray normalization image set. Use Adobe Photoshop software to perform contact surface defect segmentation on the contact multi-modal gray normalization image set. First, import the normalized image into Photoshop and use the "Quick Selection Tool" to manually select the defect area. By adjusting the sensitivity parameter of the tool (for example, the sensitivity is set to 30%), the defect area in the image can be more accurately identified. After selection, use the "Refine Edge" function to further optimize the selected area, smooth the boundary and remove noise. Finally, generate a contact surface defect segmentation marking map, where the defect area is marked in white and the background is filled in black. Perform a closed operation on the contact surface defect segmentation marking map in MATLAB. Use the bwmorph function in the MATLAB image processing toolbox and select the "close" operation to perform a closed operation on the segmentation marking map. The closed operation is a morphological operation that fills small holes and broken parts in the segmentation marking map by first dilating and then eroding. For example, set the structuring element as a 3×3 square (generated using strel('square', 3)) and perform a closed operation on the segmentation marking map. After the closed operation, a contact surface defect closed segmentation map is obtained, and the boundary of the defect area is more complete, eliminating the existing segmentation gaps. Use MATLAB software to extract the contact spiral chain code sequence from the contact surface defect closed segmentation map. First, load the closed segmentation map, and then call the bwboundaries function to extract the boundary of the defect area. This function returns a boundary matrix representing the contour of the defect area. Use the bwtraceboundary function in MATLAB to extract the spiral chain code sequence from the boundary matrix. Set the starting point as the first pixel of the boundary matrix, the search direction as clockwise (using the parameter 'W' to indicate starting the search in the west direction), and the step size as 1 pixel. Finally, generate a contact surface defect spiral chain code set. According to the contact surface defect spiral chain code set and the laser scanning path, use MATLAB for defect neighborhood dynamic expansion. First, load the spiral chain code set and the laser scanning path data. In MATLAB, use the polybuffer function to generate an extended area of the defect neighborhood based on the defect boundary represented by the spiral chain code set, with the buffer width set to 5 pixels (adjusted according to actual needs).Then, the extended area is superimposed and analyzed with the laser scanning path to ensure that the extended area covers the key areas in the laser scanning path. Finally, a mask map of the contact surface defect extension is generated. Based on the multi-modal gray-scale normalized image set of the contact and the mask map of the contact surface defect extension, the directional gradient field is calculated using MATLAB. First, the normalized image is superimposed with the extended mask map, and only the image information within the defect extension area is retained. Then, the gradient function in MATLAB is called to calculate the first-order partial derivatives of the image in the x and y directions respectively, obtaining the height gradient field. By setting the spacing parameter of the gradient function to 1 pixel. According to the directional information of the gradient field, the weights of the gradient values are adjusted so that the directional gradient field can more accurately reflect the height change of the defect area. Finally, the directional gradient distribution map of the contact surface is obtained. Using MATLAB to solve the anisotropic second-order partial derivatives of the directional gradient distribution map of the contact surface, a curvature distribution map of the contact surface is generated. After loading the directional gradient distribution map, the del2 function in MATLAB is used to calculate the anisotropic second-order partial derivatives of the image. By setting the spacing parameter of the del2 function to 1 pixel. This function returns the Laplacian operator result of the image, reflecting the curvature information of each point in the image. The calculation results are normalized to make their range between [0,1]. Finally, a curvature distribution map of the contact surface is generated, clearly showing the curvature change of the contact surface. Based on the curvature distribution map of the contact surface and the directional gradient distribution map of the contact surface, MATLAB is used to generate the deformation intensity distribution data of the contact surface. First, the curvature distribution map and the directional gradient distribution map are loaded, and then the two are multiplied pixel by pixel to obtain the preliminary data of the deformation intensity. By setting the weight factors (for example, the curvature weight is 0.6 and the gradient weight is 0.4), the influences of curvature and gradient are weighted to more accurately reflect the deformation intensity. The finally obtained deformation intensity distribution data can comprehensively reflect the deformation characteristics of the contact surface. The deformation intensity distribution data of the contact surface is weighted and fused with the mask map of the contact surface defect extension, and MATLAB is used to generate a deformation distribution map of the contact surface. After loading the deformation intensity distribution data and the defect extension mask map, the deformation intensity data is multiplied pixel by pixel with the mask map, and only the deformation information within the defect extension area is retained. By setting the fusion weight (for example, the deformation intensity weight is 0.8 and the mask weight is 0.2), the deformation characteristics display in the defect area is enhanced. The finally generated deformation distribution map of the contact surface can intuitively show the deformation situation of the defect area. Non-maximum suppression is performed on the deformation distribution map of the contact surface, and a deformation characteristic map of the contact surface is obtained using MATLAB. After loading the deformation distribution map, the imregionalmax function in MATLAB is called, and the neighborhood size is set to 3×3 pixels to perform non-maximum suppression on the deformation distribution map. This function will retain the local maximum values while suppressing other non-maximum points, thus accurately shaping the deformation characteristics and removing noise and redundant information.The finally obtained deformation feature map of the contact surface can more clearly highlight the deformation features. Perform skeleton extraction on the deformation feature map of the contact surface, and use MATLAB to generate the deformation distribution map of the contact surface. After loading the feature map, call the bwmorph function of MATLAB and select the "skel" operation for skeleton extraction. By setting the number of iterations to infinity, ensure that the thinnest skeleton is extracted. The finally generated deformation distribution map of the contact surface shows the deformation features in the form of a skeleton.

[0130] Preferably, step S2 includes the following steps:

[0131] Step S21: Perform Gaussian blur on the deformation distribution map of the contact surface according to a preset scale to generate the geometric deformation data of the contact surface at the micro scale, the geometric deformation data of the contact surface at the meso scale, and the geometric deformation data of the contact surface at the macro scale, where the preset scale includes the micro scale, the meso scale, and the macro scale;

[0132] Step S22: Calculate the mean and standard deviation of the local geometric deformation index of the contact surface at each scale;

[0133] Step S23: Map the geometric deformation data of the contact surface at the micro scale, the geometric deformation data of the contact surface at the meso scale, and the geometric deformation data of the contact surface at the macro scale to the RGB three channels respectively according to the mean and standard deviation, and perform fusion to generate the multi-scale DGDI pseudo-color map of the contact surface.

[0134] Among them, the specific calculation formulas for calculating the mean and standard deviation of the local geometric deformation index of the contact surface are as follows:

[0135]

[0136] Among them, μ scale is the arithmetic mean of the geometric deformation index of the contact surface within the current scale window, N is the total number of pixels within the current analysis window, and σ scale is the standard deviation of the geometric deformation index of the contact surface within the current scale window;

[0137] Specifically, the MATLAB software can be used to perform multi-scale Gaussian blur processing on the deformation distribution map of the contact surface, and the Gaussian blur parameters at the micro scale (standard deviation 1 pixel), the meso scale (standard deviation 3 pixels), and the macro scale (standard deviation 5 pixels) are set respectively. Different scales of geometric deformation data are generated by calling the imgaussfilt function. Then, define a 10×10 pixel analysis window, slide this window on the deformation data at each scale, and calculate the geometric deformation index D of all pixels within the window GD(x,y) mean and standard deviation. Finally, the deformation data of different scales are respectively mapped to the three RGB channels (microscopic is mapped to the red channel, mesoscopic is mapped to the green channel, and macroscopic is mapped to the blue channel). The mat2gray function is used to normalize the data to the range of [0,1], and the transparency parameter is adjusted according to the standard deviation to enhance the visualization effect of the deformation characteristics, and finally generate the multi-scale DGDI pseudo-color map of the contact

[0138] Preferably, generating the multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact in step S3 specifically includes:

[0139] Obtain the surface curvature distribution map of the contact;

[0140] Based on the surface curvature distribution map of the contact, identify the deformation gradient direction angle of the contact. A vector arrow is generated through the deformation gradient direction angle of the contact and the geometric deformation index of the contact surface. The length represents the deformation intensity, and the direction represents the deformation trend, generating the deformation gradient vector field on the contact surface. Among them, the specific calculation formula of the deformation gradient direction angle of the contact is as follows:

[0141]

[0142] where θ(x,y) is the deformation gradient direction angle, is the first-order partial derivative of the surface height function in the x direction, is the first-order partial derivative of the surface height function in the y direction;

[0143] Based on the preset unified space coordinate system, perform spatial vector superposition on the contact surface deformation distribution map, the multi-scale DGDI pseudo-color map of the contact and the deformation gradient vector field on the contact surface to generate the multi-scale deformation gradient map of the contact.

[0144] Specifically, the MATLAB software can be used to load the surface curvature distribution map of the contact from the previously processed contact surface deformation analysis data. This distribution map is obtained by calculating the second-order partial derivative of the contact surface height function z, and can clearly show the local curvature of the contact surface at each position. Use MATLAB to process the surface curvature distribution map of the contact to identify the deformation gradient direction angle of the contact. Use the gradient function to calculate the first-order partial derivatives and of the contact surface height function z in the x direction and y direction. Then, according to the formula Calculate the deformation gradient direction angle θ(x,y) of each pixel point. This direction angle represents the main direction of surface deformation. To visualize these direction angles, vector arrows are generated, where the length of the arrow represents the deformation intensity and the direction represents the deformation trend. Finally, a contact surface deformation gradient vector field is generated, clearly showing the distribution of the deformation direction and intensity. Use MATLAB software to perform spatial vector superposition on the contact surface deformation distribution map, the contact multi-scale DGDI pseudo-color map, and the contact surface deformation gradient vector field. First, define a unified spatial coordinate system to ensure that all images and data are aligned in the same coordinate framework. By calling the imshow function, the contact surface deformation distribution map and the contact multi-scale DGDI pseudo-color map are superimposed and displayed, and the quiver function is used to plot the deformation gradient vector field on the same image. Adjust the display scale of the vector field to match the brightness and color changes of the pseudo-color map, thus generating a contact multi-scale deformation gradient map.

[0145] Preferably, in step S3, constructing the contact deformation displacement field based on the contact multi-scale deformation gradient map is specifically as follows:

[0146] Extract the sub-pixel level texture features in the contact multi-scale deformation gradient map to generate a contact local binary pattern coding map;

[0147] Perform mapping fusion on the contact multi-scale deformation gradient map and the contact local binary pattern coding map to obtain a contact microscopic deformation feature map;

[0148] Obtain the initial morphology reference map of the high-voltage circuit breaker contact;

[0149] Based on the initial morphology reference map of the high-voltage circuit breaker contact, identify the deformation feature points in the contact microscopic deformation feature map to obtain a set of contact deformation feature difference coordinates;

[0150] Based on the initial morphology reference map of the high-voltage circuit breaker contact and the contact microscopic deformation feature map, perform displacement field inversion on the set of contact deformation feature difference coordinates to obtain the contact deformation displacement field.

[0151] Specifically, MATLAB software can be used to process the multi-scale deformation gradient map of the contact to extract sub-pixel level texture features. The gray-level co-occurrence matrix of the image is calculated using the graycomatrix function in MATLAB's image processing toolbox to extract the texture features of the image. Four directions (0°, 45°, 90°, 135°) and three distances (1, 2, 3 pixels) are selected to calculate the gray-level co-occurrence matrix. Then, by calculating the energy, contrast, and correlation of the gray-level co-occurrence matrix, sub-pixel level texture features are extracted. These features are encoded into binary patterns to generate the local binary pattern coding map of the contact. MATLAB software is used to perform mapping fusion on the multi-scale deformation gradient map of the contact and the local binary pattern coding map of the contact. The multi-scale deformation gradient map and the local binary pattern coding map are loaded into MATLAB, and it is ensured that the two images have the same size and resolution. Then, the imfuse function is used to fuse the two images. During the fusion process, the "add" mode is selected, and the pixel values of the two images are added. By adjusting the contrast and brightness of the fused image, the microscopic deformation feature map of the contact is generated, which clearly shows the microscopic deformation features on the surface of the contact. The initial morphology reference map of the contact is obtained from the technical documents provided by the high-voltage circuit breaker manufacturer. This reference map is scanned immediately after the contact is manufactured by a 3D scanner (such as Konica Minolta Vivid 910) and can accurately reflect the geometric morphology of the contact in the initial state. The reference map is imported into MATLAB software and preprocessed, including denoising and normalization operations. By aligning the reference map with the microscopic deformation feature map of the contact in the same coordinate system. This reference map will serve as a reference standard for subsequent deformation feature point identification and displacement field inversion. MATLAB software is used to identify the deformation feature points on the microscopic deformation feature map of the contact. First, the microscopic deformation feature map of the contact is aligned with the initial morphology reference map of the contact to ensure that the two images are completely consistent in spatial position. Then, the absolute difference between the two images is calculated using the imabsdiff function in MATLAB's image processing toolbox to generate the difference map. In the difference map, the deformation feature points appear as obvious bright regions. By setting a threshold (for example, selecting the region where the pixel value in the difference map is greater than 50), the deformation feature points are extracted, and the coordinate positions of these points are recorded to generate the contact deformation feature difference coordinate set. MATLAB software is used to perform displacement field inversion on the contact deformation feature difference coordinate set. First, the initial morphology reference map of the contact and the microscopic deformation feature map of the contact are loaded into MATLAB, and it is ensured that the two images are completely aligned in spatial position. Then, using the fitgeotrans function in MATLAB's image processing toolbox, the geometric transformation of the microscopic deformation feature map of the contact is fitted with the initial morphology reference map of the contact as a reference. By calculating the corresponding position of each deformation feature point in the reference map, the displacement vector of each feature point is obtained.Finally, the displacement vectors of all feature points are plotted on one graph to generate the contact deformation displacement field.

[0152] Particularly importantly, the construction of the contact hierarchical stress transfer model described in step S3 includes the following steps:

[0153] Collect the measured stress-strain curve of the contact material;

[0154] Construct the Ramberg-Osgood constitutive model of the high-voltage circuit breaker:

[0155]

[0156] where ∈ is the degree of material deformation, σ is the stress tensor, E is the elastic modulus, K is the hardening coefficient, and n is the strain hardening index;

[0157] Optimize the parameters of the Ramberg-Osgood constitutive model by the least square method according to the measured stress-strain curve to obtain the non-linear constitutive parameter set of the contact material;

[0158] Construct the stress transfer model according to the preset scale and the Ramberg-Osgood constitutive model to obtain the contact hierarchical stress transfer model, where the contact hierarchical stress transfer model is specifically:

[0159] Microscopic scale: Ramberg-Osgood model;

[0160] Mesoscopic scale:

[0161] Macroscopic scale: ∈ total = ∈ micro + ∈ meso + ∈ macro ;

[0162] where ∈ is the total strain, E is the elastic modulus, K is the hardening coefficient, n is the strain hardening index, C slip is the interface slip correction coefficient, ∈ total is the total strain, ∈ micro is the microscopic scale strain, ∈ meso is the mesoscopic scale strain, ∈ macro is the macroscopic scale strain.

[0163] Specifically, a universal material testing machine (such as Instron 5982) can be used in a material mechanics laboratory to perform a tensile test on the high-voltage circuit breaker contact material to obtain its stress-strain curve. The contact material is made into a standard tensile specimen and installed in the fixture of the testing machine. By gradually applying tensile force, the deformation of the specimen under different loads is recorded, including stress (stress tensor) and strain (degree of deformation). During the test, a strain gauge (such as Vishay Micro-Measurements strain gauge) is used to monitor the strain change of the specimen in real time, and the stress and strain data are recorded through the data acquisition system of the testing machine. Finally, the measured stress-strain curve of the contact material is obtained. The collected stress-strain curve of the contact material is analyzed using MATLAB software, and the Ramberg-Osgood constitutive model is constructed. First, the measured stress-strain data is imported into MATLAB, and according to the Ramberg-Osgood model formula: Where ∈ is the degree of deformation of the material, σ is the stress tensor, E is the elastic modulus, K is the hardening coefficient, and n is the strain hardening exponent. According to the known elastic modulus E of the material (for example, for a certain steel, E = 210GPa), and assuming the initial hardening coefficient K and strain hardening exponent n values (such as K = 100MPa, n = 0.15), the theoretical stress-strain curve is preliminarily calculated through the model. The parameters of the Ramberg-Osgood constitutive model are optimized using MATLAB software. The lsqcurvefit function in MATLAB is used to optimize the hardening coefficient K and strain hardening exponent n in the model with the measured stress-strain curve as the target. The initial parameter range of the optimization is set (such as K ranges from 50 to 200MPa, n ranges from 0.1 to 0.3), and the error between the model curve and the measured curve is minimized through multiple iterative calculations. Finally, the optimized nonlinear constitutive parameter set is obtained, including the accurate K and n values. According to the optimized Ramberg-Osgood constitutive model parameters, the contact layered stress transfer model is constructed using MATLAB software. The three scales of micro, meso and macro are preset, and the corresponding stress transfer model is defined according to the characteristics of each scale. The Ramberg-Osgood model is directly used at the micro scale; the interface slip effect is considered at the meso scale, and the model is: Among them, C slip is the interface slip correction coefficient, which is set to 0.05 according to the grain boundary characteristics of the material. The macroscopic scale superimposes the strains of the microscopic and mesoscopic scales to obtain the total strain: total =∈ micro +∈ meso +∈ macro ; Through the above steps, a hierarchical model is constructed that can reflect the stress transfer and deformation behavior of contact materials at different scales.

[0164] Preferably, step S4 includes the following steps:

[0165] Step S41: Invert the contact surface strain tensor through finite element numerical simulation based on the contact deformation displacement field and the contact delamination stress transfer model;

[0166] Step S42: Solve the contact surface strain tensor through the contact delamination stress transfer model to obtain the contact multi-scale stress field data;

[0167] Step S43: Fuse the contact surface strain tensor and the contact multi-scale stress field data to obtain the contact material response characteristic map;

[0168] Step S44: Based on the contact material response characteristic map and the contact delamination stress transfer model, conduct contact motion simulation on the high-voltage circuit breaker contact to obtain the contact dynamic stress field data;

[0169] Step S45: Construct a contact non-linear deformation response model based on the contact dynamic stress field data;

[0170] Step S46: Collect the actual working condition parameters of the contact; according to the actual working condition parameters of the contact, conduct contact deformation simulation on the contact non-linear deformation response model to obtain the contact deformation co-evolution data.

[0171] Wherein, the inversion of the contact surface strain tensor through finite element numerical simulation is specifically:

[0172]

[0173] Where ∈ ij is the contact surface strain tensor, u i is the component of the displacement vector in the i direction, x j is the coordinate axis in the j direction, u j is the component of the displacement vector in the j direction, x i is the coordinate axis in the i direction;

[0174] Specifically, the ANSYS finite element analysis software can be used to conduct numerical simulation on the contact. First, import the contact deformation displacement field into

[0175] ANSYS, and set the material properties and boundary conditions according to the contact delamination stress transfer model. Mesh the contact model by defining the mesh element type (such as tetrahedral element) and the mesh density (about 10 elements per millimeter). Use the inversion function of ANSYS to calculate the contact surface strain tensor ∈ ij , and the formula is Where u i and u jThey are the components of the displacement vector in the i and j directions respectively. By setting appropriate solution parameters (such as the convergence criterion is 1 e-6 ), run the simulation and obtain the distribution results of the contact surface strain tensor. Use ANSYS software to solve the contact surface strain tensor. According to the previously optimized contact layer stress transfer model parameters (such as elastic modulus E, hardening coefficient K, strain hardening index n), set the nonlinear constitutive relationship of the material in ANSYS. By defining the loading path (such as the gradually applied mechanical load) and boundary conditions (such as fixed constraints), run the nonlinear static analysis. ANSYS automatically solves the multi-scale stress field data of the contact according to the input strain tensor and material model, including the stress distribution at the micro, meso and macro scales. Use MATLAB software to fuse the contact surface strain tensor and multi-scale stress field data. First, import the strain tensor and stress field data exported by ANSYS into MATLAB and ensure that they are completely aligned in spatial position. Write a script to perform point-by-point multiplication or addition operations on the strain tensor and stress field data to generate the contact material response characteristic map. Adjust the contrast and brightness of the image, and represent different response intensities as different colors through color mapping. Finally, the generated contact material response characteristic map intuitively shows the mechanical response characteristics of the contact at different positions. Import the contact material response characteristic map into ANSYS and set the material properties and boundary conditions in combination with the contact layer stress transfer model. Define the initial position and motion trajectory of the contact (such as the displacement and speed of the opening and closing operation), and simulate the motion process of the contact by setting the dynamic analysis module (such as transient dynamics analysis). During the simulation, consider the interaction between the contact and the surrounding environment (such as friction and electromagnetic force), and extract the contact dynamic stress field data. These data include the stress distribution and changes of the contact at different time points. Use MATLAB software to analyze the contact dynamic stress field data and construct a contact nonlinear deformation response model. According to the dynamic stress field data and the actual motion trajectory of the contact, write a script to define the model formula where u is the dynamic deformation displacement vector, t is the time, is the stress gradient tensor, T is the temperature field, and E is the electromagnetic field parameter. By fitting the dynamic stress field data and the contact motion trajectory, the parameter relationships in the model are determined. For example, the nonlinear relationship between stress and displacement is obtained by least squares fitting, and the effects of temperature and electromagnetic field on deformation are considered. Finally, a nonlinear deformation response model that can describe the dynamic deformation behavior of the contact under complex working conditions is obtained. The actual working condition parameters of the contact are collected in the actual operating environment of the high-voltage circuit breaker. High-precision sensors (such as strain gauges, temperature sensors, and electromagnetic field sensors) are installed on the contact to monitor the stress, temperature, and electromagnetic field changes of the contact in real time. The collected actual working condition parameters are imported into the MATLAB software, and deformation simulation is carried out in combination with the previously constructed contact nonlinear deformation response model. By setting the parameters in the model (such as stress, temperature, and electromagnetic field) as the actual monitored values, the simulation is run to obtain the deformation co-evolution data of the contact under actual working conditions. These data include the deformation displacement and stress distribution of the contact at different time points.

[0176] Preferably, step S5 includes the following steps:

[0177] Step S51: Detect and process the outliers in the contact deformation co-evolution data to obtain the standard deformation evolution data;

[0178] Step S52: Extract the eigenvectors from the standard deformation evolution data to obtain the contact motion trajectory eigenvectors;

[0179] Step S53: Obtain the initial simple 3D model of the contact; based on the initial simple 3D model of the contact, perform coordinate transformation and parameter mapping on the contact motion trajectory eigenvectors to obtain the contact motion parameter mapping space;

[0180] Step S54: Build a correlation model for the contact motion parameter mapping space to obtain the contact motion parameter correlation data;

[0181] Step S55: Screen the key motion parameters from the contact motion trajectory eigenvectors according to the contact motion parameter correlation data to obtain the contact motion parameter candidate set;

[0182] Step S56: Build a motion trajectory model based on the initial simple 3D model of the contact and the contact motion parameter candidate set to obtain the contact motion trajectory model.

[0183] Specifically, the Python programming language and its scientific computing libraries NumPy and SciPy are used to detect and process outliers in the co-evolution data of contact deformation. First, the data is loaded, and the Matplotlib library is used to plot the data distribution graph to visualize the outliers in the data. The Z-score-based outlier detection method is adopted to calculate the Z-score value of each data point. According to experience, data points with an absolute Z-score greater than 3 are regarded as outliers. The stats.zscore function of SciPy is used to calculate the Z-score, and the outliers are filtered out through boolean indexing. Finally, the obtained standard deformation evolution data is saved as a new dataset. Python and its machine learning library scikit-learn are used to extract eigenvectors from the standard deformation evolution data. The dataset is loaded into a Pandas DataFrame for easy data manipulation. The PCA class (Principal Component Analysis) of scikit-learn is used to reduce the dimensionality of the data. The number of principal components is set to 3 to extract the three most important eigenvectors from the data. By calling the fit_transform method, the original data is transformed into a new feature space to obtain the eigenvectors of the contact movement trajectory. The SolidWorks software is used to obtain the initial simple 3D model of the contact. First, the model file is opened in SolidWorks, and the coordinate system settings of the model are checked. The "Transform" function of SolidWorks is used to align the model coordinate system with the coordinate system of the actual contact. The eigenvectors of the contact movement trajectory are imported into SolidWorks, and the eigenvectors are mapped to the coordinate system of the model by writing a VBA script. The coordinate transformation matrix is defined in the script to transform the eigenvectors from the original coordinate system to the model coordinate system. Finally, the mapping space of the contact movement parameters is obtained. Python and its data visualization libraries Matplotlib and Seaborn are used to build a correlation model for the mapping space of the contact movement parameters. First, the data in the mapping space is loaded into a Pandas DataFrame, and the correlation coefficient matrix between the parameters is calculated. The heatmap function of Seaborn is used to plot the correlation coefficient heatmap to visually display the correlation between the parameters. The LinearRegression class of scikit-learn is used to build a linear regression model to analyze the linear relationship between the parameters. Through model training and verification, the correlation data of the contact movement parameters is obtained. Python and its machine learning library scikit-learn are used to screen the key movement parameters from the correlation data of the contact movement parameters. The correlation data is loaded, and the SelectKBest class combined with the f_regression scoring function is used to evaluate the importance of each parameter. The number of selected parameters is set to 5 to ensure that the most critical parameters are screened out. Through model training and verification, the candidate set of the contact movement parameters is obtained.Use SolidWorks Motion to perform motion trajectory modeling on the initial simple 3D model of the contact and the candidate set of contact motion parameters. First, open the initial simple 3D model in SolidWorks and import the candidate set parameters into the model. Utilize the motion analysis function of SolidWorks Motion to set the motion constraints and driving conditions. By defining parameters such as the initial position, velocity, and acceleration of the motion trajectory, run the simulation analysis to generate the contact motion trajectory model.

[0184] Particularly importantly, step S56 further includes the following steps:

[0185] Step S561: Evaluate the error propagation of the candidate set of contact motion parameters to obtain the contact motion trajectory error index;

[0186] Step S562: Iteratively correct the candidate set of contact motion parameters based on the contact motion trajectory error index to obtain a candidate solution for contact trajectory correction;

[0187] Step S563: Verify the consistency of the candidate solution for contact trajectory correction to obtain contact trajectory correction data;

[0188] Step S564: Construct a preliminary model for motion parameter reconstruction based on the contact trajectory correction data;

[0189] Step S565: Extract the constraint conditions of the preliminary model for motion parameter reconstruction to obtain the contact motion constraint feature set;

[0190] Step S566: Perform motion trajectory modeling based on the initial simple 3D model of the contact and the contact motion constraint feature set to obtain the contact motion trajectory model.

[0191] Specifically, the Python programming language and its scientific computing libraries NumPy and SciPy are used to evaluate the error propagation of the candidate set of contact motion parameters. First, the candidate set data is loaded, and the error propagation model is constructed using the matrix operation function of NumPy. The error propagation formula is defined, considering the covariance matrix and Jacobian matrix between parameters to quantify the impact of parameter uncertainty on the motion trajectory. Through multiple Monte Carlo simulations, a large number of sample data are generated, and the error index of each sample is calculated. Finally, the error indexes of the contact motion trajectory are obtained, including position error, velocity error, and acceleration error. Python and its optimization library SciPy are used to iteratively correct the candidate set of contact motion parameters. First, an objective function is defined, which evaluates the quality of the parameter set based on the motion trajectory error index. The minimize function of SciPy is selected, and the BFGS optimization algorithm is used to iteratively optimize the parameter set. The initial parameter values are set, and the convergence criterion (such as the gradient norm being less than) is defined. Through multiple iterations, the error index is gradually reduced, and finally, a candidate solution for contact trajectory correction is obtained. The MATLAB software is used to verify the consistency of the candidate solution for contact trajectory correction. First, the candidate solution is imported into MATLAB and compared with the historical data set. The corrcoef function of MATLAB is used to calculate the correlation coefficient between the candidate solution and the historical data to evaluate its consistency. A comparison graph of the candidate solution and the historical data is plotted, and the plot function is used to visually display the degree of coincidence between the two. Finally, the corrected data for the contact trajectory is obtained. Python and its machine learning library scikit-learn are used to construct a preliminary model for reconstructing motion parameters based on the corrected data of the contact trajectory. The corrected data is loaded into a Pandas DataFrame and preprocessed, including normalization and filling of missing values. The RandomForestRegressor class of scikit-learn is selected to construct a regression model, and the random forest algorithm is used to fit the parameter relationship in the corrected data. Through cross-validation, the performance of the model is evaluated, and the hyperparameters (such as the number of trees and the maximum depth) are adjusted. Finally, a preliminary model for reconstructing motion parameters is obtained, which can predict the motion parameters of the contact under different working conditions. The SolidWorks software is used to extract the constraint conditions of the preliminary model for reconstructing motion parameters. First, the initial simple 3D model of the contact is opened in SolidWorks, and the parameters of the preliminary model are imported into the model. The "Constraint Analysis" function of SolidWorks is used to identify the physical and geometric constraints in the model. Through analysis, the key constraint conditions are extracted, including joint limits, collision detection, and motion range limits. These constraint conditions are organized into a feature set of contact motion constraints. The SolidWorks Motion is used to model the motion trajectory of the initial simple 3D model of the contact and the feature set of contact motion constraints. The initial simple 3D model is opened in SolidWorks, and the constraint feature set is applied to the model.Set motion constraint conditions, such as joint types and motion ranges, and define driving conditions, such as speed and acceleration. By running a simulation analysis, a contact motion trajectory model is generated, including the position, speed, and acceleration data of the contact at different time points.

[0192] Preferably, the present invention also provides a modeling system for a power device, which is used to execute the modeling method for a power device as described above. The modeling system for a power device includes:

[0193] An image registration module, which is used to collect a set of three-dimensional tomography images of the contact and construct an initial simple three-dimensional model of the contact; perform mutual information maximization registration on the set of three-dimensional tomography images of the contact to obtain a contact registration image data set; identify the deformation of the contact of the high-voltage circuit breaker based on the contact registration image data set, and generate a contact surface deformation distribution map;

[0194] A deformation analysis module, which is used to construct a multi-scale DGDI pseudo-color map of the contact based on the contact surface deformation distribution map;

[0195] A model construction module, which is used to generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact; construct a contact deformation displacement field based on the multi-scale deformation gradient map of the contact; construct a contact layered stress transfer model;

[0196] A response simulation module, which is used to perform contact material response analysis based on the contact deformation displacement field and the contact layered stress transfer model to obtain a contact material response characteristic map; perform contact deformation simulation based on the contact material response characteristic map and the contact layered stress transfer model to obtain contact deformation co-evolution data;

[0197] A contact motion modeling module, which is used to perform motion trajectory mapping on the initial simple three-dimensional model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; perform motion simulation on the initial simple three-dimensional model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

[0198] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application document are intended to be included in the present invention.

[0199] The above description is only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A modeling method for power equipment, characterized in that, Applied to the moving contact of a high-voltage circuit breaker, the modeling method for power equipment includes the following steps: Step S1: Collect a set of three-dimensional tomography images of the contact and construct an initial simple three-dimensional model of the contact; perform mutual information maximization registration on the set of three-dimensional tomography images of the contact to obtain a dataset of registered contact images; identify the deformation of the contact of the high-voltage circuit breaker based on the dataset of registered contact images, and generate a distribution map of the surface deformation of the contact; Step S2: Construct a multi-scale DGDI pseudo-color map of the contact based on the distribution map of the surface deformation of the contact; Step S3: Generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact; construct a deformation displacement field of the contact based on the multi-scale deformation gradient map of the contact; construct a hierarchical stress transfer model of the contact; Step S4: Perform contact material response analysis based on the deformation displacement field of the contact and the hierarchical stress transfer model of the contact to obtain a contact material response characteristic map; perform contact deformation simulation based on the contact material response characteristic map and the hierarchical stress transfer model of the contact to obtain contact deformation co-evolution data; Step S5: Perform motion trajectory mapping on the initial simple three-dimensional model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; perform motion simulation on the initial simple three-dimensional model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

2. The modeling method for power equipment according to claim 1, wherein In Step S1, the collection of the set of three-dimensional tomography images of the contact is specifically as follows: Use a laser interferometer to measure the high-voltage circuit breaker contact at multiple positions to obtain a set of geometric parameters of the high-voltage circuit breaker contact; Construct an initial simple three-dimensional model of the contact based on the set of geometric parameters of the high-voltage circuit breaker contact; Determine the laser scanning path based on the initial simple three-dimensional model of the contact; Perform multi-angle layer scanning on the high-voltage circuit breaker contact according to the laser scanning path to obtain a set of three-dimensional tomography images of the contact, where the multi-angle layer scanning includes optical coherence tomography imaging at azimuths of 0°, 45°, 90°, 135°, and 180°, and the set of three-dimensional tomography images of the contact includes tomography slice data with a transverse resolution of 3 μm and a longitudinal resolution of 1 μm of the contact surface topography.

3. The modeling method for power equipment according to claim 1, wherein In Step S1, the mutual information maximization registration of the set of three-dimensional tomography images of the contact is specifically as follows: Calculate the mutual information values between images at different angles in the set of three-dimensional tomography images of the contact, where the specific calculation formula is as follows; Where I(A,B) is the mutual information value between the reference image A and the image B to be registered, p(a,b) is the joint probability distribution of the gray values of the two images at the same spatial position, p(a) is the marginal probability distribution of the reference image A, p(b) is the marginal probability distribution of the image B to be registered, a is the gray value of the reference image A, and b is the gray value of the image B to be registered; Maximize the mutual information value to obtain the maximized mutual information; where the maximization is specifically as follows: By adjusting the translation parameters (t ′ x , t ′ y , t ′ z ) and rotation parameters (θ x , θ y , θ z ) of the image B to be registered; The parameter update formula for adjusting the image B to be registered is: where θ is the parameter vector to be optimized, including the translation parameters (t ′ x , t ′ y , t ′ z ) and the rotation parameters (θ x , θ y , θ z ). t ′ x , t ′ y , t ′ z are the translation amounts of the image in the X, Y, and Z axis directions respectively, and θ x , θ y , θ z are the rotation angles of the image around the X, Y, and Z axes respectively. α is the step size for controlling the parameter update, is the gradient of the mutual information I(A, B) with respect to the parameter θ; Apply θ to the image B to be registered new Perform a spatial transformation to obtain the aligned image B' Superimpose the reference image A and all the aligned images B' according to the spatial position to generate a dataset of registered contact images.

4. The modeling method for power equipment according to claim 1, characterized in that In Step S1, the identification of the deformation of the contact of the high-voltage circuit breaker based on the dataset of registered contact images is specifically as follows: Perform multi-modal gray normalization on the dataset of registered contact images to obtain a set of multi-modal gray-normalized contact images; Perform contact surface defect segmentation on the multi-modal gray-scale normalized image set of the contact to obtain a contact surface defect segmentation label map; Perform dilation processing on the defect regions in the contact surface defect segmentation label map to obtain a contact surface defect dilation map; perform a closing operation on the defect regions of the contact surface defect dilation map to obtain a contact surface defect closed segmentation map; Extract the contact spiral chain code sequence based on the contact surface defect closed segmentation map to generate a contact surface defect spiral chain code set; Perform defect neighborhood dynamic expansion according to the contact surface defect spiral chain code set and the laser scanning path to generate a contact surface defect expansion mask map; Calculate the directional gradient field based on the multi-modal gray-scale normalized image set of the contact and the contact surface defect expansion mask map to obtain a contact surface directional gradient distribution map; Calculate the contact surface deformation curvature based on the contact surface directional gradient distribution map, and generate a contact surface curvature distribution map according to the contact surface deformation curvature; Generate contact surface deformation intensity distribution data based on the contact surface curvature distribution map and the contact surface directional gradient distribution map; Perform weighted fusion of the contact surface deformation intensity distribution data and the contact surface defect expansion mask map to generate a contact surface deformation distribution map; Perform non-maximum suppression on the contact surface deformation distribution map to obtain a contact surface deformation feature map; Perform skeleton extraction on the contact surface deformation feature map to generate a contact surface deformation distribution map.

5. The modeling method for power equipment according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Perform Gaussian blur on the contact surface deformation distribution map according to a preset scale to generate contact micro-scale geometric deformation data, contact medium-scale geometric deformation data, and contact macro-scale geometric deformation data, where the preset scale includes micro-scale, medium-scale, and macro-scale; Step S22: Calculate the mean and standard deviation of the local contact surface geometric deformation index at each scale; Step S23: Map the contact micro-scale geometric deformation data, contact medium-scale geometric deformation data, and contact macro-scale geometric deformation data to the RGB three channels respectively according to the mean and standard deviation, and perform fusion to generate a contact multi-scale DGDI pseudo-color map.

6. The modeling method for power equipment according to claim 1, wherein In step S3, generating the contact multi-scale deformation gradient map based on the contact multi-scale DGDI pseudo-color map is specifically: Obtain the contact surface curvature distribution map; Identify the contact deformation gradient direction angle based on the contact surface curvature distribution map, generate a vector arrow through the contact deformation gradient direction angle and the contact surface geometric deformation index, where the length represents the deformation intensity and the direction represents the deformation trend, to generate a contact surface deformation gradient vector field, and the specific calculation formula of the contact deformation gradient direction angle is as follows: where θ(x, y) is the direction angle of the deformation gradient, is the first-order partial derivative of the surface height function in the x direction, is the first-order partial derivative of the surface height function in the y direction; Perform spatial vector superposition on the contact surface deformation distribution map, the contact multi-scale DGDI pseudo-color map, and the contact surface deformation gradient vector field based on a preset unified space coordinate system to generate a contact multi-scale deformation gradient map.

7. The modeling method for power equipment according to claim 1, wherein In step S3, constructing the contact deformation displacement field based on the contact multi-scale deformation gradient map is specifically: Extract the sub-pixel level texture features in the contact multi-scale deformation gradient map to generate a contact local binary pattern coding map; Map and fuse the multi-scale deformation gradient map of the contact and the local binary pattern coding map of the contact to obtain the microscopic deformation feature map of the contact; Obtain the initial morphology reference map of the high-voltage circuit breaker contact; Based on the initial morphology reference map of the high-voltage circuit breaker contact, identify the deformation feature points of the microscopic deformation feature map of the contact to obtain the contact deformation feature difference coordinate set; Based on the initial morphology reference map of the high-voltage circuit breaker contact and the microscopic deformation feature map of the contact, perform displacement field inversion on the contact deformation feature difference coordinate set to obtain the contact deformation displacement field.

8. The modeling method for power equipment according to claim 1, wherein Step S4 includes the following steps: Step S41: Invert the surface strain tensor of the contact through finite element numerical simulation based on the contact deformation displacement field and the contact hierarchical stress transfer model; Step S42: Solve the surface strain tensor of the contact through the contact hierarchical stress transfer model to obtain the multi-scale stress field data of the contact; Step S43: Fuse the surface strain tensor of the contact and the multi-scale stress field data of the contact to obtain the contact material response feature map; Step S44: Based on the contact material response feature map and the contact hierarchical stress transfer model, perform contact motion simulation on the high-voltage circuit breaker contact to obtain the contact dynamic stress field data; Step S45: Based on the contact dynamic stress field data, construct a contact non-linear deformation response model; Step S46: Collect the actual working condition parameters of the contact; according to the actual working condition parameters of the contact, perform contact deformation simulation on the contact non-linear deformation response model to obtain the contact deformation co-evolution data.

9. The modeling method for power equipment according to claim 1, wherein Step S5 includes the following steps: Step S51: Detect and process the outliers in the contact deformation co-evolution data to obtain the standard deformation evolution data; Step S52: Extract the eigenvectors from the standard deformation evolution data to obtain the contact motion trajectory eigenvector; Step S53: Obtain the initial simple 3D model of the contact; based on the initial simple 3D model of the contact, perform coordinate transformation and parameter mapping on the contact motion trajectory eigenvector to obtain the contact motion parameter mapping space; Step S54: Perform correlation modeling on the contact motion parameter mapping space to obtain the contact motion correlation data; Step S55: According to the contact motion correlation data, screen the key motion parameters of the contact motion trajectory eigenvector to obtain the contact motion parameter candidate set; Step S56: Import the contact motion parameter candidate set into the initial simple 3D model of the contact and perform motion trajectory modeling to obtain the contact motion trajectory model.

10. A modeling system for power equipment, characterized in that, For implementing the modeling method for power equipment as described in claim 1, the modeling system for power equipment includes: An image registration module, configured to collect a set of 3D tomography images of the contact and construct an initial simple 3D model of the contact; perform mutual information maximization registration on the set of 3D tomography images of the contact to obtain a contact registration image data set; identify the contact deformation of the high-voltage circuit breaker based on the contact registration image data set, and generate a contact surface deformation distribution map; A deformation analysis module, configured to construct a multi-scale DGDI pseudocolor map of the contact based on the contact surface deformation distribution map; A model construction module, which is used to generate a multi-scale deformation gradient map of the contact based on the multi-scale DGDI pseudo-color map of the contact; construct a deformation displacement field of the contact based on the multi-scale deformation gradient map of the contact; construct a hierarchical stress transfer model of the contact; A response simulation module, which is used to perform contact material response analysis based on the contact deformation displacement field and the contact hierarchical stress transfer model to obtain a contact material response characteristic map; perform contact deformation simulation based on the contact material response characteristic map and the contact hierarchical stress transfer model to obtain contact deformation co-evolution data; A contact motion modeling module, which is used to perform motion trajectory mapping on the initial simple three-dimensional model of the contact according to the contact deformation co-evolution data to obtain a set of contact motion parameters; perform motion simulation on the initial simple three-dimensional model of the contact based on the set of contact motion parameters to obtain a contact motion trajectory model.

Images