Power equipment state simulation and real-time deduction reconstruction method and system

Abstract

The invention discloses a power equipment state simulation and real-time deduction reconstruction method and system, and belongs to the technical field of power equipment state monitoring and emulation.The method comprises the steps that firstly, internal physical field sample data of equipment under different working conditions are obtained through off-line numerical simulation; secondly, on the basis of the sample data, constructing a rapid deduction model of a function relationship between an input variable and a response value by utilizing a radial basis function neural network; then, acquiring real-time monitoring data of equipment operation and inputting the real-time monitoring data into the deduction model; then, physical field data values of all grid nodes in the equipment are calculated in real time through the model; and finally, through voxelization, adjacent unit search and trilinear interpolation, accurately mapping the physical field data to the three-dimensional model of the equipment, generating a visual cloud picture, and realizing real-time and visual reconstruction of the state of the equipment. The technical problems that traditional monitoring is incomplete and the simulation speed is low are solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power equipment state monitoring and simulation, and particularly relates to a power equipment state simulation and real-time deduction reconstruction method and system. BACKGROUND

[0002] In a power system, the stable operation of power equipment is crucial. Accurately grasping the internal state of power equipment and timely discovering potential fault hazards are of great significance to the safe and reliable operation of the power system.

[0003] Traditional power equipment state monitoring methods mainly rely on sensors to obtain part of the operating parameters of the equipment. However, due to the complexity and diversity of the internal physical fields of power equipment, it is difficult to provide complete internal state information of the equipment through sensor measurement or simulation means alone. Therefore, it is urgent to integrate computational analysis data and sensing measurement data to support accurate evaluation of the state of the equipment.

[0004] Existing monitoring means mainly rely on sensors installed on the surface or key nodes of the equipment (such as the top oil temperature sensor of an oil-immersed transformer and the gas density sensor of SF6 gas insulated equipment). These sensors can only obtain the operating parameters of the local and surface layers of the equipment and cannot penetrate the complex structure of the equipment (such as the transformer core, the winding of the reactor, and the internal part of the GIS basin insulator) to obtain the physical field information of the core area. For example, the hot spot temperature of the transformer winding is a key indicator for evaluating the insulation aging rate. However, traditional sensors can only measure the oil temperature and cannot directly monitor the internal hot spot of the winding, making it difficult for maintenance personnel to accurately judge the degree of insulation degradation. At the same time, there is a lack of direct correlation between sensor data and changes in the internal physical field of the equipment, and the internal state cannot be deduced from the surface parameters, which easily leads to the dilemma of "data redundancy but information missing".

[0005] Although existing multi-physical field simulation technology can simulate and calculate the internal physical field of power equipment, it is usually based on offline numerical simulation methods, which take a long time to calculate and cannot meet the real-time state monitoring requirements. To make up for the limitations of sensor monitoring, the industry often uses numerical simulation techniques such as the finite element method and the finite volume method to calculate the internal physical field of the equipment. However, such methods have significant drawbacks. On the one hand, offline simulation requires the construction of a refined grid model (such as a transformer three-dimensional model with millions of grid points), and a single simulation takes tens of minutes to hours (such as the traditional finite element method for calculating the thermal field of a transformer, which takes more than 46 minutes), which cannot respond to changes in the operating conditions of the equipment (such as sudden load rate increases and environmental temperature changes) in real time. On the other hand, the simulation model parameters (such as the thermal conductivity of the material and the heat dissipation boundary conditions) are mostly set based on design values, which deviate from the actual operating state of the equipment, and cannot be dynamically corrected by integrating real-time sensor data, resulting in a disconnection between the simulation results and the actual state, making it difficult to directly use for fault diagnosis and state evaluation.

[0006] In addition, how to effectively associate and visually display the simulation calculation results with the actual state of the equipment is also a difficult problem currently faced. The existing simulation outputs are mostly discrete grid node data or two-dimensional cloud maps, which cannot be accurately mapped with the three-dimensional entity model of the equipment, and the internal physical field distribution law is difficult for the operation and maintenance personnel to intuitively understand. At the same time, there is a lack of technical means for converting the real-time calculated physical field data into visualized state, and it is impossible to quickly locate the potential fault area such as local hot spot and stress concentration, resulting in the phenomenon that "the calculation results cannot effectively support the operation and maintenance decision" existing universally.

[0007] Therefore, there is an urgent need for a method capable of real-time and accurate deduction of the internal physical field of the power equipment and reconstruction of the equipment state in an intuitive way to meet the high requirements of modern power systems on the state monitoring of power equipment. SUMMARY

[0008] The technical problem to be solved by the present application is to provide a power equipment state simulation and real-time deduction reconstruction method and system to solve the technical problems of incomplete internal state perception, untimely evaluation and difficult state visualization of power equipment caused by the limitation of sensing information and the lag of offline simulation in the prior art, realize real-time deduction calculation and intuitive state reconstruction of the internal physical field of the power equipment, and improve the accuracy and timeliness of the state evaluation of the power equipment.

[0009] The present application adopts the following technical solutions: A power equipment state simulation and real-time deduction reconstruction method, comprising the following steps: S1, obtaining data samples of the internal physical field of the power equipment through numerical simulation calculation; S2, based on the data samples, constructing a function relationship between input variables and response values of sampling points in the three-dimensional space of the power equipment, and constructing a deduction model of the internal physical field of the power equipment by using a radial basis neural network, wherein the input variables include structural parameters, operating parameters and environmental parameters of the power equipment, and the response values are data values of each internal physical field of the power equipment; S3, obtaining real-time data in the operation of the power equipment, and inputting the real-time data into the deduction model, wherein the real-time data include state monitoring sensing data, operating data and environmental data; S4, using the deduction model to calculate the physical field data values corresponding to each grid node in the power equipment in real time; S5, mapping the calculated physical field data values corresponding to each grid node in the power equipment with the three-dimensional model of the power equipment to realize real-time reconstruction of the state of the power equipment.

[0010] Preferably, in step S1, the data samples of the internal physical field of the power equipment obtained through numerical simulation calculation comprise: A physical model is constructed based on the geometric parameters of the power equipment, and basic material parameters are configured for the physical model. The power equipment entity is divided into hexahedral meshes using multiphysics control mesh units, and sampling points are designed. A space of influencing factors and variables covering the power equipment under different operating conditions is set. Data of the physical field of the power equipment is obtained through numerical calculation, forming the data sample.

[0011] Preferably, in step S2, the core function of constructing the deduction model of the internal physical field of the power equipment using a radial basis function neural network is formally expressed as:

[0012] in, yes With the sample points Euclidean distance in a spatial region These are basis functions. It is the Euclidean norm. It is a positive constant or zero. Each basis function The weight, .

[0013] Preferably, in step S3, acquiring real-time data during the operation of the power equipment specifically involves: By installing various condition monitoring sensors on the power equipment, the condition monitoring sensor data, operation data and environmental data of the power equipment during operation are collected as the real-time data.

[0014] Preferably, in step S4, the physical field data values ​​corresponding to each grid node inside the power equipment are calculated in real time using the simulation model, specifically as follows: During the simulation calculation process, the input parameters of the simulation model are dynamically adjusted according to the changes in the real-time data described in step S3. Then, the physical field data values ​​corresponding to each grid node inside the power equipment are calculated through the simulation model after parameter adjustment. The physical field data values ​​include electromagnetic field data values, thermal field data values ​​and structural stress field data values. The calculated physical field data values ​​are stored in the result data file.

[0015] Preferably, the overall error of the physical field data values ​​corresponding to each grid node inside the power equipment is within ±1°C.

[0016] Preferably, in step S5, the step of mapping the calculated physical field data values ​​corresponding to the internal grid nodes of the power equipment with the node data of the three-dimensional model of the power equipment to realize real-time reconstruction of the power equipment state includes: S501. Voxelize the 3D model of the equipment and divide the model space into... × × Uniform grid, , , Use positive integers to generate the center point coordinates of each voxel. ; S502, for the center point coordinates of each voxel Locate the finite element hexahedron containing the element and read the coordinates of its eight vertices from the result data file obtained in step S4. and the corresponding physical quantity values , ; S503. The physical quantity values ​​of the 8 vertices are weighted and summed, and the physical quantity values ​​of the voxel points are calculated using the trilinear interpolation method. S504. After normalizing the physical quantity values ​​of each voxel point obtained in step S503, map them to the color space to generate a cloud map of the voxel mesh, thus completing the real-time reconstruction of the power equipment status.

[0017] Preferably, the center point coordinates for:

[0018] in, These are the 3D models of the equipment in x , y , z The maximum boundary point on the axis, , , These are the 3D models of the equipment in x , y , z Minimum boundary point on the axis, , , , These are the intervals between the center points of two adjacent voxels on the x, y, and z axes, respectively.

[0019] Preferably, voxel points physical quantity value for:

[0020] in, voxel points Relative distance on the x, y, z axes It is the weight value.

[0021] Preferably, the weight value The specific expression is as follows:

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028] .

[0029] In a second aspect, the embodiment of the present application provides a power equipment state simulation and real-time deduction reconstruction system, comprising: a data module, which obtains data samples of internal physical fields of the power equipment through numerical simulation calculation; a construction module, which constructs a function relationship between input variables and response values of sampling points in a three-dimensional space of the power equipment based on the data samples, and constructs a deduction model of the internal physical fields of the power equipment by using a radial basis neural network, wherein the input variables include structural parameters, operation parameters and environmental parameters of the power equipment, and the response values are data values of each internal physical field of the power equipment; a real-time module, which obtains real-time data in the operation of the power equipment, and inputs the real-time data into the deduction model, wherein the real-time data include state monitoring sensing data, operation data and environmental data; a calculation module, which calculates physical field data values corresponding to each grid node in the power equipment in real time by using the deduction model; a reconstruction module, which maps the calculated physical field data values corresponding to each grid node in the power equipment to node data of a three-dimensional model of the power equipment, so as to realize real-time reconstruction of the state of the power equipment.

[0030] Preferably, in the data module, the obtaining of the data samples of the internal physical fields of the power equipment through numerical simulation calculation comprises: constructing a physical model according to geometric parameters of the power equipment, and configuring basic material parameters for the physical model; performing hexahedral mesh partitioning on an entity of the power equipment by using a multi-physical field control mesh unit, and designing sampling points at the same time; setting influencing factors and variable spaces covering different working conditions of the power equipment; obtaining data of physical fields of the power equipment through numerical calculation, so as to form the data samples.

[0031] Preferably, in the constructing module, the inferring model of the internal physical field of the power equipment is constructed by using the radial basis neural network, and a core function is expressed as:

[0032] wherein, is the Euclidean distance of the first sample point and the second sample point in a spatial region, is a base function, is a Euclidean norm, is a positive constant or zero, is a weight of each base function . .

[0033] Preferably, in the real-time module, the real-time data in the operation of the power equipment are specifically obtained as follows: The state monitoring data, operation data and environmental data of the power equipment in operation are collected by various state monitoring sensors installed on the power equipment, as the real-time data.

[0034] Preferably, in the calculating module, the physical field data values corresponding to each grid node in the power equipment are calculated in real time by using the inferring model, and the calculation specifically includes: In the inferring calculation process, the input parameters of the inferring model are dynamically adjusted according to the changes of the real-time data in the real-time module, and then the physical field data values corresponding to each grid node in the power equipment are calculated by using the inferring model after the adjustment of the parameters, the physical field data values include electromagnetic field data values, thermal field data values and structural stress field data values, and the calculated physical field data values are stored in a result data file, and the overall error of the calculated physical field data values corresponding to each grid node in the power equipment fluctuates within ±1°C.

[0035] Preferably, in the reconstructing module, the physical field data values corresponding to each grid node in the power equipment are mapped with the three-dimensional model of the power equipment by node data to realize real-time reconstruction of the state of the power equipment, and the reconstruction specifically includes: The three-dimensional model of the equipment is voxelized, and the model space is divided into uniform grids × × , , , wherein, n, m and p are positive integers, and the center point coordinates of each voxel are generated . The center point coordinates are as follows:

[0036] wherein, are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , , are the intervals between the center points of two adjacent voxels on the x, y, z axes respectively; the center point coordinate of each voxel is , the finite element hexahedron unit in which the center point is located is found, and the coordinates of the 8 vertices of the finite element hexahedron unit and the corresponding physical quantity values are read from the obtained result data file , , ; the physical quantity values of the 8 vertices are weighted and summed, the physical quantity value of the voxel point is calculated by using the trilinear interpolation method, and the physical quantity value of the voxel point is:

[0037] wherein, are the relative distances of the voxel point on the x, y, z axes respectively, is a weight value, and the specific expression of the weight value is as follows:

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044] ; After the obtained physical quantity values of each voxel point are normalized, the physical quantity values are mapped to a color space to generate a cloud chart of the voxel grid, and the real-time reconstruction of the state of the power equipment is completed. ​

[0045] In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the power equipment state simulation and real-time deduction reconstruction method when executing the computer program.

[0046] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium including a computer program, wherein the computer program implements the steps of the power equipment state simulation and real-time deduction reconstruction method when executed by a processor.

[0047] In a fifth aspect, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the power equipment state simulation and real-time deduction reconstruction method when executing the computer program.

[0048] In a sixth aspect, an embodiment of the present application provides an electronic device including a computer program, wherein the computer program implements the steps of the power equipment state simulation and real-time deduction reconstruction method when executed by the electronic device.

[0049] Compared with the prior art, the present application has at least the following beneficial effects: A power equipment state simulation and real-time deduction reconstruction method, by the mode of "offline training and online application", the calculation burden is preposed, so that the speed is improved by thousands of times in the real-time monitoring stage, which meets the harsh requirements of the power system on the real-time perception of the state. At the same time, this method combines the advantages of simulation and sensing data, which can not only obtain the physical field information (such as internal winding temperature and magnetic flux density) of any position inside the equipment that cannot be directly measured by sensors through simulation, but also dynamically correct the model using real-time data, so that the deduction result is closer to the real operation condition of the equipment, realizing the leap from "local perception" to "full field transparency", and providing an unprecedented data basis for accurate state evaluation and early fault warning.

[0050] Further, through standardized physical modeling and grid partitioning, the accuracy of the simulation data and the consistency with the physical structure of the real equipment are ensured. And the systematic setting of the variable space covering various possible operating conditions and fault modes makes the generated sample data able to fully represent the dynamic behavior of the equipment, greatly enhancing the generalization ability and robustness of the constructed radial basis neural network deduction model. This means that the trained model is not only suitable for the current operating condition, but also can make reliable deduction prediction for various load changes, environmental fluctuations and even early abnormal states that may occur in the future, avoiding the risk of model failure in unknown operating conditions.

[0051] Further, the characteristics of local response make it have the advantages of fast convergence speed and strong approximation ability when fitting the spatial distribution of complex physical fields. By establishing the mathematical form, the "neural network" of black box is transformed into a "derivation model" with clear mathematical and physical meaning, enhancing the interpretability and repeatability of the scheme. The clear training process ensures that the model can accurately learn the complex nonlinear mapping relationship between the input variables (structure, operation, environmental parameters) and the output response (full-field physical quantities) from high-fidelity simulation data, thereby ensuring that in real-time inference, it can reproduce the accuracy comparable to the original simulation with extremely high computational efficiency.

[0052] Further, a multi-dimensional and comprehensive real-time input system is constructed. State monitoring sensor data (such as oil temperature, vibration) directly reflects the running health status of the equipment; operation data (such as load current, voltage) are the core excitation that drives the change of internal physical fields of the equipment; environmental data (such as ambient temperature, humidity) directly affect the heat dissipation and insulation performance of the equipment. Taking these three categories of data as model inputs together, the inference model can comprehensively consider all key factors affecting the state of the equipment, dynamically and adaptively adjust the inference results, ensuring the accuracy and real-time of state reconstruction, making the inference system an intelligent system that can breathe and perceive external changes, rather than a static lookup table.

[0053] Further, directly aiming at the core demand of power equipment state evaluation. The thermal field (temperature distribution) is the key to evaluate the insulation aging and load capacity of the equipment; the electromagnetic field (electric field, magnetic field intensity) is related to the insulation design, loss and electromagnetic compatibility of the equipment; the structural stress field directly affects the mechanical life and safety of the equipment. By inferring these core physical fields, the invention can reveal the key state information inside the equipment, such as locating the hot spot of the winding, analyzing the local overheating caused by leakage flux, evaluating the mechanical stress under short-circuit current, etc., thereby providing accurate data support for deep state evaluation and fault diagnosis based on multi-physical field coupling, and realizing the sublimation from "macroscopic operation parameter monitoring" to "microscopic physical field analysis".

[0054] Further, it realizes high-fidelity and high-precision conversion from "data" to "visualization". Voxelization discretizes the continuous three-dimensional space into uniform grids, laying the foundation for visualization. By searching for adjacent units and trilinear interpolation, this method can fully utilize the existing high-precision finite element calculation results, ensuring the spatial continuity and minimal accuracy loss of physical field data during voxelization. This fine mapping relationship enables the final generated visualization cloud map to realistically and smoothly reflect the gradient changes and local features (such as hot spots, high stress areas) of the internal physical fields of the equipment, achieving a "millimeter-level spatial resolution" display effect, greatly improving the intuitiveness and precise positioning ability of state display, and providing a "CT scan" like internal perspective view of the equipment for operation and maintenance personnel.

[0055] Further, the trilinear interpolation is a standard method of three-dimensional space interpolation, and the shape function ensures the linear characteristics of the physical quantity change in the hexahedral element. The weight calculation formula is clear, so that the technical solution can be clearly and accurately implemented and reproduced. This method is stable and efficient, and can ensure that the physical quantity value calculated at the voxel point is the smooth weighted average of the eight vertex values of the element, thereby generating a visually continuous and non-mutated cloud chart, accurately revealing the distribution rule of the physical field inside the device, and is a key technology for realizing high-quality state reconstruction.

[0056] Further, the modular design makes the system structure clear, and each module has clear responsibilities, which is beneficial to the development, integration, maintenance and function expansion of the system. Each module corresponds to a key stage of the method, ensuring the full-process automation and collaborative work from data preparation, model construction, real-time acquisition, fast calculation to final visualization. The existence of the system enables the above method to be deployed as a complete product or solution in actual substations or monitoring centers, and converts technical innovation into actual productivity.

[0057] It can be understood that the beneficial effects of the above-mentioned second aspect to the sixth aspect can be referred to the related description in the first aspect, which will not be repeated here.

[0058] In summary, the present application realizes the leap of power equipment state monitoring from local perception and offline analysis to full-field transparency and real-time deduction; through the radial basis neural network proxy model, the calculation speed is improved by thousands of times, meeting the real-time requirement; through three-dimensional data mapping, the internal physical field is reconstructed with millimeter-level resolution, which significantly improves the accuracy, timeliness and visualization level of state evaluation, and provides core technical support for intelligent operation and maintenance.

[0059] The technical solutions of the present application will be further described in detail below with the aid of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 The method flowchart of the present application; Figure 2 The internal physical field deduction model construction flowchart of the device of the present application; Figure 3 The transformer internal temperature field deduction calculation result cloud chart example; Figure 4 The schematic diagram of the computer device provided by an embodiment of the present application; Figure 5 The block diagram of a chip provided by an embodiment of the present application.

[0061] Wherein, 60. Computer device; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access storage unit; 6202. Cache storage unit; 6203. Read-only storage unit; 6204. Program / utility; 6205. Program modules; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0063] In the description of the present application, it should be understood that the terms "comprise" and "include" indicate the existence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or sets thereof.

[0064] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, unless otherwise clearly indicated by the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0065] It should be further understood that the term "and / or" used in the specification and the appended claims of the present application means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations, for example, A and / or B can represent three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present application generally represents an "or" relationship between the front and rear associated objects.

[0066] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present application to describe the preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, the first preset range can also be referred to as the second preset range, and similarly, the second preset range can also be referred to as the first preset range without departing from the scope of the embodiments of the present application.

[0067] Depending on the context, the word "if' can be interpreted to mean "when" or "upon" or "in response to determining" or "in response to detecting." Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]" or "in response to detecting [the stated condition or event]."

[0068] Various structural diagrams according to the disclosed embodiments of the present application are shown in the drawings. These diagrams are not drawn to scale, in which certain details are exaggerated for clarity and others omitted. The shapes and relative sizes of the various regions, layers, and the relative positions of these regions / layers shown in the drawings are merely examples and can deviate in practice due to manufacturing tolerances or technical limitations, and regions / layers with different shapes, sizes, and relative positions can be additionally designed according to actual needs by those skilled in the art.

[0069] The present application provides a power equipment state simulation and real-time deduction reconstruction method, physical field data samples are obtained through numerical simulation, a deduction model is constructed using a radial basis neural network, internal grid node physical field data values are calculated by inputting real-time data of the equipment, and state reconstruction is realized by mapping with a three-dimensional model; the system includes corresponding functional modules, and the implementation method is landed. The internal multi-physical field deduction of the power equipment is driven based on the perception data, the calculation speed of the multi-physical field simulation model is improved, the physical field data values of the internal grid nodes of the equipment calculated are mapped with the three-dimensional model of the equipment, and the real-time reconstruction of the equipment state is realized. The problems of incomplete traditional monitoring and non-real-time simulation are solved, and the accuracy and timeliness of the equipment state evaluation are improved.

[0070] Referring to Figure 1 The present application provides a power equipment state simulation and real-time deduction reconstruction method, physical field data samples are obtained through numerical simulation, a deduction model is constructed using a radial basis neural network, internal grid node physical field data values are calculated by inputting real-time data of the equipment, and state reconstruction is realized by mapping with a three-dimensional model; the system includes corresponding functional modules, and the implementation method is landed. The internal multi-physical field deduction of the power equipment is driven based on the perception data, the calculation speed of the multi-physical field simulation model is improved, the physical field data values of the internal grid nodes of the equipment calculated are mapped with the three-dimensional model of the equipment, and the real-time reconstruction of the equipment state is realized. The problems of incomplete traditional monitoring and non-real-time simulation are solved, and the accuracy and timeliness of the equipment state evaluation are improved. S1, obtaining equipment internal physical field data samples through numerical simulation calculation; S101, performing physical modeling according to the geometric parameters of the power equipment, and setting basic material parameters; S102, performing hexahedral mesh partitioning on the equipment entity using a multi-physical field control grid unit, and designing sampling points; S103, setting influencing factors and variable spaces covering different working conditions of the equipment; S104, obtaining equipment physical field data through numerical calculation to form sample data.

[0071] S2, constructing a function relationship between input variables and response values of the sampling points in the three-dimensional space of the equipment based on the data samples, and constructing a deduction model using a radial basis neural network; Referring toFigure 2 Based on data samples, a functional relationship between input variables and response values ​​of sampling points in the three-dimensional space of the device is constructed. The input variables include the structural parameters, operating parameters, and environmental parameters of the device, and the response values ​​are the data values ​​of various physical fields inside the device.

[0072] A deductive model is constructed using a radial basis function neural network. The core function of this model can be formally expressed as:

[0073] in, , , yes With the sample points Euclidean distance in a spatial region These are basis functions. It is the Euclidean norm. It is a positive constant or zero. Each basis function The weight.

[0074] By learning and training on sample data, the weights and parameters of the neural network are determined, and the parameters are continuously tuned until the inference model passes the accuracy evaluation.

[0075] S3. Acquire status monitoring sensor data, operation data, environmental data, etc. of power equipment during operation, and transmit the real-time data to the simulation model; By installing various condition monitoring sensors on power equipment, real-time data such as condition monitoring sensor data, operation data, and environmental data during equipment operation are acquired, and the real-time data is passed to the model as input variables for the inference model.

[0076] S4. Calculate the physical field data values ​​corresponding to each grid node inside the device in real time using the simulation model; During the simulation calculation, the input parameters of the simulation model are dynamically adjusted according to the changes in the real-time data in step S3. The simulation model is used to calculate the physical field data values ​​corresponding to each grid node inside the device in real time, including electromagnetic field, thermal field, structural stress field, etc., and store them in the result data file.

[0077] S5. Map the calculated physical field data values ​​corresponding to the internal mesh nodes of the equipment to the three-dimensional model of the equipment to realize real-time reconstruction of the equipment status.

[0078] S501. Voxelize the 3D model of the equipment and divide the model space into... × × A uniform grid, in which , , Use positive integers to generate the center point coordinates of each voxel. ;

[0079]

[0080]

[0081]

[0082] in, These are the 3D models of the equipment in x, y, z The maximum boundary point on the axis, , , These are the 3D models of the equipment in x, y, z The minimum boundary point on the axis. .

[0083] S502. Search for neighboring cells, corresponding to the coordinates of each voxel. Locate the finite element hexahedron containing the element, and read the coordinates of the eight vertices of the hexahedron from the result data file obtained in step S4. and the corresponding physical quantity values ; S503. Calculate the physical quantity values ​​of voxel points by weighted summation of the physical quantities of the 8 vertices and using trilinear interpolation.

[0084]

[0085] in, voxel points The relative distances on the x, y, and z axes are weighted by the relative distances from the center point to each dimension, specifically:

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093] S504, mapping the physical quantity of each voxel point obtained in step S503 to a color space after normalization to generate a cloud chart of the voxel grid, and realizing state reconstruction of the device.

[0094] In another embodiment of the present application, a power device state simulation and real-time deduction reconstruction system is provided, which can be used to realize the power device state simulation and real-time deduction reconstruction method described above. Specifically, the power device state simulation and real-time deduction reconstruction system comprises a data module, a construction module, a real-time module, a calculation module, and a reconstruction module.

[0095] The data module obtains data samples of the internal physical field of the power device through numerical simulation calculation. The construction module constructs a functional relationship between input variables and response values of sampling points in the three-dimensional space of the power device based on the data samples, and constructs a deduction model of the internal physical field of the power device using a radial basis neural network, wherein the input variables include structural parameters, operating parameters, and environmental parameters of the power device, and the response values are data values of each internal physical field of the power device. The real-time module obtains real-time data in the operation of the power device, and inputs the real-time data into the deduction model, wherein the real-time data includes state monitoring sensor data, operating data, and environmental data. The calculation module uses the deduction model to calculate the physical field data values corresponding to each grid node inside the power device in real time. The reconstruction module maps the calculated physical field data values corresponding to each grid node inside the power device to the node data of the three-dimensional model of the power device, to realize real-time reconstruction of the state of the power device.

[0096] In the data module, the data samples of the internal physical field of the power device obtained through numerical simulation calculation include: A physical model is constructed according to the geometric parameters of the power device, and basic material parameters are configured for the physical model. Hexahedral grid subdivision is performed on the power device entity using a multi-physical field control grid element, and sampling points are designed. Influencing factors and variable spaces covering different operating conditions of the power device are set. Data of the physical field of the power device is obtained through numerical calculation to form the data samples.

[0097] In the construction module, the deduction model of the internal physical field of the power device constructed using the radial basis neural network has a core function formalized as:

[0098] wherein, is and the first sample point Euclidean distance in spatial region, is a base function, is a Euclidean norm, is a positive constant or zero, is a weight of each base function , .

[0099] In the real-time module, the real-time data in the operation of the power equipment are specifically: Through various state monitoring sensors installed on the power equipment, state monitoring data, operation data and environmental data during the operation of the power equipment are collected as the real-time data.

[0100] In the calculation module, the physical field data values corresponding to each grid node inside the power equipment are calculated in real time by using the deduction model, specifically: During the deduction calculation process, the input parameters of the deduction model are dynamically adjusted according to the changes of the real-time data in the real-time module, and then the physical field data values corresponding to each grid node inside the power equipment are calculated by the deduction model after adjusting the parameters, the physical field data values include electromagnetic field data values, thermal field data values and structural stress field data values, and the calculated physical field data values are stored in a result data file, the overall error of the calculated physical field data values corresponding to each grid node inside the power equipment fluctuates within ±1°C.

[0101] In the reconstruction module, the calculated physical field data values corresponding to each grid node inside the power equipment are mapped with the node data of the three-dimensional model of the power equipment to realize real-time reconstruction of the state of the power equipment, including: The three-dimensional model of the equipment is voxelized, and the model space is divided into × × uniform grids, , , are positive integers, and the center point coordinates of each voxel are generated; The center point coordinates are:

[0102] wherein, are the maximum boundary points of the three-dimensional model of the equipment on x , y , z axis, , , are the maximum boundary points of the three-dimensional model of the equipment on x , y , zMinimum boundary point on the axis, , , , These represent the intervals between the center points of two adjacent voxels on the x, y, and z axes, respectively; expressed by the formula:

[0103]

[0104]

[0105] For the center point coordinates of each voxel Locate the finite element hexahedron containing the element, and read the coordinates of the eight vertices of that finite element hexahedron from the resulting data file. and the corresponding physical quantity values , ; The physical quantity values ​​of the eight vertices are weighted and summed, and the physical quantity values ​​of the voxel points are calculated using trilinear interpolation. physical quantity value for:

[0106] in, voxel points Relative distance on the x, y, z axes It's the weight value, the weight value The specific expression is as follows:

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113] ; After normalizing the physical quantity values ​​of each voxel point, they are mapped to the color space to generate a cloud map of the voxel mesh, thus completing the real-time reconstruction of the power equipment status.

[0114] The application provides a terminal device, which comprises a processor and a memory, the memory is used for storing a computer program, the computer program comprises program instructions, and the processor is used for executing the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components and the like, which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement a corresponding method flow or a corresponding function; the processor in the embodiment of the application can be used for the operation of the power equipment state simulation and real-time deduction reconstruction method, which comprises: obtaining data samples of internal physical fields of the power equipment through numerical simulation calculation; based on the data samples, a function relationship between input variables and response values of sampling points in a three-dimensional space of the power equipment is constructed, a deduction model of the internal physical fields of the power equipment is constructed by using a radial basis neural network, wherein the input variables comprise structural parameters, operation parameters and environmental parameters of the power equipment, and the response values are data values of each internal physical field of the power equipment; real-time data in the operation of the power equipment is obtained, and the real-time data is input into the deduction model, wherein the real-time data comprises state monitoring sensing data, operation data and environmental data; the deduction model is used to calculate physical field data values corresponding to each grid node in the power equipment in real time; the calculated physical field data values corresponding to the grid nodes in the power equipment are mapped with node data of a three-dimensional model of the power equipment, so as to realize real-time reconstruction of the state of the power equipment.

[0115] Please refer to Figure 4The terminal device is a computer device. The computer device 60 in this embodiment includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and capable of running on the processor 61. When the computer program 63 is executed by the processor 61, the method for estimating the concentration of radioactive iodine species in the post-accident containment in the embodiment is implemented. To avoid repetition, details are not described here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of the models / units in the power equipment state simulation and real-time deduction reconstruction system in the embodiment are implemented. To avoid repetition, details are not described here.

[0116] The computer device 60 can be a desktop computer, a notebook computer, a palm computer, a cloud server, or the like. The computer device 60 can include, but is not limited to, the processor 61 and the memory 62. Those skilled in the art can understand that the computer device 60 can include more or fewer components, or some components can be combined, or different components can be included, for example, the computer device can also include an input / output device, a network access device, a bus, and the like. Figure 4 The computer device 60 is only an example and does not constitute a limitation on the computer device 60, and can include more or fewer components than those shown, or some components can be combined, or different components can be included, for example, the computer device can also include an input / output device, a network access device, a bus, and the like.

[0117] The processor 61 can be a central processing unit (CPU), and can also be other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, or the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0118] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or a memory of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, or the like.

[0119] Further, the memory 62 can include both a volatile memory unit and a nonvolatile memory unit. The memory 62 can also include a removable memory unit. The memory 62 is utilized by the computer device 60 to store data and programs and is further utilized to temporarily store output data.

[0120] Referring to Figure 5 , the terminal device is an electronic device 600 in the form of a general purpose computing device. Components of the electronic device can include, but are not limited to, at least one processing unit 610, at least one memory unit 620, a bus 630 that connects the various platform components including the memory unit 620 and the processing unit 610, a display unit 640, and the like.

[0121] The memory unit stores program code that can be executed by the processing unit 610 such that the processing unit 610 performs the steps described in the above method section of this specification in accordance with the various example embodiments of this application. For example, the processing unit 610 can perform the steps shown in Figure 1 .

[0122] The memory unit 620 can include a read-only memory (ROM) 6203 in the form of a nonvolatile memory unit, as well as a random access memory (RAM) 6201 and / or a cache memory unit 6202 in the form of a volatile memory unit.

[0123] The memory unit 620 can further include a program / utility 6204 having a set of program modules 6205, including but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination thereof, which may

[0124] The bus 630 can be representative of one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration bus, a processor or local bus using any of a variety of bus structures, and the like.

[0125] The electronic device 600 can also communicate with one or more external devices 700 such as a keyboard or pointing device, a Bluetooth device, or a database, and / or one or more devices that enable a user to interact with the electronic device 600 and / or one or more devices (e.g., routers, modems) that enable the electronic device 600 to communicate with one or more other computing devices. Such communication can occur via an input / output interface 650. Still yet, the electronic device 600 can communicate with one or more networks, such as a local area network (LAN), a general wide area network (WAN), and / or the Internet, through a network adapter 660. The network adapter 660 can be any of a plurality of different types of adapters suitable for interfacing the electronic device 600 to various networking schemes. It should be appreciated that, while not shown explicitly, other hardware and / or software modules could be used in conjunction with the electronic device 600, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0126] Embodiment 4 The present application also provides a storage medium, specifically a computer readable storage medium, which is a memory device in the terminal device, and is used to store programs and data. It should be understood that the computer readable storage medium herein can include the built-in storage medium in the terminal device, and of course can also include the expansion storage medium supported by the terminal device, and can be any tangible medium containing or storing programs, which can be used by or in conjunction with an instruction execution system, device or apparatus. The computer readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that more specific examples of the computer readable storage medium include an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0127] Computer readable storage media further includes data signals transported through a carrier wave and a communications medium, including any medium that now exists or is developed in the future that transmits program code in the form of packets, computer files, or data structures. Program code manifested as data signals includes any read-only instructions, instructions that are written to memory, and instructions that are executed by a processing unit. Computer readable storage media do not logically include carrier waves but can also include media that stores or transmits program code in the form of packets, computer files, or data structures.

[0128] Program code used by or in connection with the described embodiments can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network or a wide area network, or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider.

[0129] The one or more instructions stored in the computer readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the power equipment state simulation and real-time reconstruction method in the above embodiments; the one or more instructions stored in the computer readable storage medium are loaded and executed by the processor to perform the following steps: Data samples of internal physical fields of the power equipment are obtained through numerical simulation calculation; based on the data samples, a function relationship between input variables and response values of sampling points in a three-dimensional space of the power equipment is constructed, and a reconstruction model of the internal physical fields of the power equipment is constructed by using a radial basis neural network, wherein the input variables include structural parameters, operating parameters and environmental parameters of the power equipment, and the response values are data values of the internal physical fields of the power equipment; real-time data in the operation of the power equipment are obtained, and the real-time data are input into the reconstruction model, wherein the real-time data include state monitoring sensing data, operating data and environmental data; the reconstruction model is used to calculate the physical field data values corresponding to each grid node in the power equipment in real time; the calculated physical field data values corresponding to each grid node in the power equipment are mapped with the node data of the three-dimensional model of the power equipment, so as to realize real-time reconstruction of the state of the power equipment.

[0130] The database involved in each embodiment provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, and the like, without limitation. The processor involved in each embodiment provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, and the like, without limitation.

[0131] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. All other embodiments obtained by a person of ordinary skill in the art without creative work based on the embodiments in the present application belong to the scope of protection of the present application.

[0132] Taking an oil-immersed transformer as an example, the following steps are included: S1, physical modeling is performed according to the geometric parameters of the transformer, and basic material parameters are set; a multi-physical field control grid element is used to perform hexahedral mesh division on the transformer, and a Latin hypercube design method is used to construct a sample point data set; influencing factors and variable spaces of the transformer under different working conditions are set, wherein the electrical working condition part includes that the high-voltage winding bears a direct-current bias current range of 2-6 A, and the step is set to 1 A, different load working conditions set the load rate at 50%-140%, and the step is 10%; the thermal fault working condition considers the influence of environmental temperature change on oil circulation and heat exchange, and sets the environmental temperature change range to 5°C-40°C, and the step is 5°C. The data of the internal temperature field of the transformer is obtained by numerical calculation to form a data sample, as shown in Figure 3 .

[0133] S2, based on the data sample, a function relationship between the input variables and response values of the sampling points in the internal three-dimensional space of the transformer is constructed, and a radial basis neural network is used to construct a deduction model. In order to quantitatively evaluate the calculation results of the model, the root mean square error and the maximum error are used for result evaluation.

[0134] S3, state monitoring sensor data, operation data, environmental data, and the like in the operation of the power equipment are obtained, and real-time data is transmitted to the deduction model, for example, the environmental temperature is 20°C, the load rate is 70%, and the bias current is 3A. S4. Calculate the temperature field data values ​​corresponding to each grid node inside the transformer in real time using the deduction model; S5. Voxelize the 3D model of the transformer, for example, by dividing the model space into... × × A uniform grid is generated, producing 1,000,000 voxels; neighboring cells are searched, corresponding to the coordinates of each voxel. Locate the finite element hexahedron containing the element, and read the coordinates of the eight vertices of the hexahedron from the result data file obtained in step S4. and the corresponding physical quantity values The physical quantities of the voxel points are calculated by weighted summation of the physical quantities of the eight vertices and trilinear interpolation. The obtained physical quantities of each voxel point are normalized and mapped to the color space to generate a cloud map of the voxel mesh, thereby realizing the state reconstruction of the device.

[0135] Assume the boundary extent of the transformer model is as follows (unit: meters). =0.0, =2.0, =-1.0, =1.0, =0.0, =1.5 Then the grid spacing:

[0136]

[0137]

[0138] First voxel point ,

[0139] That is, the coordinates of the first voxel point are (0.0001, -0.9999, 0.000075); intermediate point ,

[0140] That is, the coordinates of the intermediate element point are (0.0101, -0.9899, ​​0.007575). The coordinates of other voxel points are calculated in a similar manner.

[0141] The physical quantity values ​​of voxel points are calculated as follows:

[0142] in, voxel points The relative distances on the x, y, z axes, the weight values are determined by the relative distances from the center point to each dimension, and are specifically:

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150] Set the voxel point The relative distances on the x, y, z axes are respectively , then the weight value = .

[0151] The present application selects the temperature field of an oil-immersed transformer as an example, compares and analyzes the simulation results and the model prediction values under typical working conditions (ambient temperature is 20°C, load rate is 70%, and magnetic current is 3A), and develops a temperature field rapid calculation model. The temperature field rapid calculation model can restore the spatial temperature distribution characteristics under the complex structure of the transformer, the prediction results are highly consistent with the original simulation, especially in the fitting effect of the temperature rise trend in the local hot spot area and the boundary area of the winding, and the model has good nonlinear fitting ability and spatial continuity expression ability. The overall error fluctuates within ±1°C, and there is no systematic deviation or large-scale deviation, indicating that the model has stable generalization performance under different thermal field gradients.

[0152] In addition to the accuracy performance, the transformer internal temperature field rapid calculation model constructed by the present application also has significant advantages in calculation efficiency. Taking the typical working condition (ambient temperature 20°C, load rate 70%, and magnetic current 3A) as an example, the traditional thermal field simulation based on the finite element method takes about 46 minutes and 32 seconds, and depends on a high-performance solver and a complex mesh division and convergence control process.

[0153] Compared with the traditional thermal field simulation, the rapid deduction model can complete the temperature prediction of about 50,000 space points in about 1.27 seconds after receiving the same input conditions (normalized excitation characteristics and spatial coordinate points), and the overall deduction time is shortened to about one thousandth of the original simulation, greatly improving the multi-condition rapid evaluation and system real-time response ability In summary, the power equipment state simulation and real-time deduction reconstruction method and system can realize real-time deduction calculation and intuitive state reconstruction of the internal physical field of the power equipment, improve the accuracy and timeliness of the power equipment state evaluation, construct a rapid deduction model through a radial basis neural network, combine real-time sensing data as parameter input of the deduction model, realize second-level dynamic update of the internal electric field, magnetic field, temperature field and other multi-physical fields of the equipment, and the calculation speed is significantly better than that of the traditional numerical simulation, and through the three-dimensional model node mapping technology, the fault point can be located to millimeter-level spatial resolution, thereby providing accurate basis for operation and maintenance decision.

[0154] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific names of each functional unit and module are only for easy distinction, and do not limit the protection scope of the application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0155] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0156] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0157] In the embodiments of the present application, it should be understood that the disclosed apparatus / terminal and method can be implemented in other manners. For example, the embodiments of the apparatus / terminal described above are merely schematic, and the division of the modules or units is merely logical function division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0158] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0159] In addition, each functional unit in the various embodiments of the present application can be integrated into a processing unit, or each unit can be a physically independent unit, or two or more units can be integrated into a unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0160] The integrated module / unit, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, all or part of the flow of the above-mentioned embodiment methods can be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer readable medium can include or exclude content according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0161] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps in one or more flow or blocks

[0162] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps in one or more flow or blocks

[0163] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps in one or more flow or blocks

[0164] The above merely provides the technical idea of the present application, and cannot be used to limit the protection scope of the present application. Any modification made according to the technical idea of the present application, on the basis of the technical solutions, falls within the protection scope of the claims of the present application.

Claims

1. A method for power equipment state simulation and real-time deduction reconstruction, characterized in that, The method comprises the following steps: S1, obtaining data samples of internal physical fields of the power equipment through numerical simulation calculation; S2, based on the data samples, constructing a function relationship between input variables and response values of sampling points in a three-dimensional space of the power equipment, and constructing a deduction model of the internal physical fields of the power equipment by using a radial basis neural network, wherein the input variables comprise structural parameters, operating parameters and environmental parameters of the power equipment, and the response values are data values of each internal physical field of the power equipment; S3, obtaining real-time data in the operation of the power equipment, and inputting the real-time data into the deduction model, wherein the real-time data comprise state monitoring sensor data, operating data and environmental data; S4, using the deduction model to calculate physical field data values corresponding to each grid node in the power equipment in real time; S5, mapping the calculated physical field data values corresponding to the grid nodes in the power equipment to a three-dimensional model of the power equipment to realize real-time reconstruction of the state of the power equipment.

2. The power equipment state simulation and real-time deduction reconstruction method according to claim 1, characterized in that, In step S1, the data samples of the internal physical fields of the power equipment are obtained through numerical simulation calculation, comprising: A physical model is constructed according to the geometric parameters of the power equipment, and basic material parameters are configured for the physical model; hexahedral mesh subdivision is performed on the entity of the power equipment by using a multi-physical field control mesh unit, and sampling points are designed; influencing factors and variable spaces covering different operating conditions of the power equipment are set; data of the physical fields of the power equipment are obtained through numerical calculation to form the data samples.

3. The power equipment state simulation and real-time deduction reconstruction method according to claim 1, characterized in that, In step S2, the deduction model of the internal physical fields of the power equipment is constructed by using a radial basis neural network, and the core function is formalized as: in, yes With the sample points Euclidean distance in a spatial region These are basis functions. It is the Euclidean norm. It is a positive constant or zero. Each basis function The weight, .

4. The power equipment state simulation and real-time deduction reconstruction method according to claim 1, characterized in that, In step S3, the real-time data in the operation of the power equipment are obtained, comprising: State monitoring sensor data, operating data and environmental data during the operation of the power equipment are collected by various state monitoring sensors installed on the power equipment as the real-time data.

5. The power equipment state simulation and real-time deduction reconstruction method according to claim 1, characterized in that, In step S4, the deduction model is used to calculate the physical field data values corresponding to each grid node in the power equipment in real time, comprising: In the deduction calculation process, the input parameters of the deduction model are dynamically adjusted according to the changes of the real-time data in step S3, and then the physical field data values corresponding to each grid node in the power equipment are calculated by the adjusted deduction model, wherein the physical field data values comprise electromagnetic field data values, thermal field data values and structural stress field data values, and the calculated physical field data values are stored in a result data file.

6. The power equipment state simulation and real-time deduction reconstruction method according to claim 5, characterized in that, The overall error of the calculated physical field data values corresponding to each grid node in the power equipment fluctuates within ±1°C.

7. The power equipment state simulation and real-time deduction reconstruction method according to claim 1, characterized in that, In step S5, the calculated physical field data values corresponding to the grid nodes in the power equipment are mapped to a three-dimensional model of the power equipment to realize real-time reconstruction of the state of the power equipment, comprising: S501, voxelizing the device three-dimensional model, dividing the model space into × × uniform grids, , , is a positive integer, generating the center point coordinates of each voxel ; S502、find the center point coordinate of each voxel , find the finite element hexahedron unit where the voxel is located, read the coordinates of the 8 vertices of the finite element hexahedron unit from the result data file obtained in step S4 and the corresponding physical quantity value , ; S503, weighted sum of physical values of 8 vertices is performed, and trilinear interpolation method is used to calculate physical values of voxel points; S504, after normalization processing of the physical values of each voxel point obtained in step S503, the physical values are mapped to a color space to generate a cloud chart of the voxel grid, and real-time reconstruction of the state of the power equipment is completed.

8. The power equipment state simulation and real-time deduction reconstruction method according to claim 7, characterized in that, The center point coordinates are: wherein, are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , , are the intervals between the center points of two adjacent voxels on the x, y, z axes respectively.

9. The power equipment state simulation and real-time deduction reconstruction method according to claim 7, characterized in that, voxel points physical quantity values are: wherein, are voxel points are relative distances in x, y, z axes, are weight values.

10. The power equipment state simulation and real-time deduction reconstruction method according to claim 9, characterized in that, weight values The specific expression of the weight values is as follows: 。 11. A power equipment state simulation and real-time inference reconstruction system, characterized in that, Comprise: A data module obtains data samples of internal physical fields of the power equipment through numerical simulation calculation; A construction module constructs a function relationship between input variables and response values of sampling points in a three-dimensional space of the power equipment based on the data samples, and constructs a deduction model of the internal physical fields of the power equipment by using a radial basis neural network, wherein the input variables include structural parameters, operating parameters and environmental parameters of the power equipment, and the response values are data values of the internal physical fields of the power equipment; A real-time module obtains real-time data in the operation of the power equipment, and inputs the real-time data into the deduction model, wherein the real-time data include state monitoring sensor data, operating data and environmental data; A calculation module calculates physical field data values corresponding to each grid node in the power equipment in real time by using the deduction model; A reconstruction module maps the calculated physical field data values corresponding to each grid node in the power equipment to a node data of a three-dimensional model of the power equipment, to realize real-time reconstruction of the state of the power equipment.

12. The power equipment state simulation and real-time deduction reconstruction system according to claim 11, characterized in that, In the data module, the data samples of the internal physical fields of the power equipment obtained through numerical simulation calculation include: A physical model is constructed according to geometric parameters of the power equipment, and basic material parameters are configured for the physical model; hexahedral grid subdivision is performed on the entity of the power equipment by using a multi-physical field control grid unit, and sampling points are designed; influencing factors and variable spaces of the power equipment under different operating conditions are set; data of the physical fields of the power equipment are obtained through numerical calculation, to form the data samples.

13. The power equipment state simulation and real-time deduction reconstruction system according to claim 11, characterized in that, In the construction module, the deduction model of the internal physical fields of the power equipment constructed by using the radial basis neural network has a core function formalized as: wherein is the Euclidean distance between the first sample point and the second sample point in the spatial region, is a basis function, is the Euclidean norm, is a positive constant or zero, is a weight of each basis function , and .

14. The power equipment state simulation and real-time deduction reconstruction system of claim 11, wherein, In the real-time module, the real-time data in the operation of the power equipment are obtained as follows: State monitoring sensor data, operating data and environmental data of the power equipment in operation are collected by various state monitoring sensors installed on the power equipment, as the real-time data.

15. The power equipment state simulation and real-time deduction reconstruction system of claim 11, wherein, In the calculation module, the physical field data values corresponding to each grid node in the power equipment are calculated in real time by using the deduction model, as follows: In the deduction calculation process, input parameters of the deduction model are dynamically adjusted according to changes in the real-time data in the real-time module, and the physical field data values corresponding to each grid node in the power equipment are calculated by using the deduction model with the adjusted parameters, wherein the physical field data values include electromagnetic field data values, thermal field data values and structural stress field data values, and the calculated physical field data values are stored in a result data file, and the overall error of the calculated physical field data values corresponding to each grid node in the power equipment fluctuates within ±1°C.

16. The power equipment state simulation and real-time deduction reconstruction system of claim 11, wherein, In the reconstruction module, the physical field data values corresponding to each grid node in the power equipment are mapped to a node data of a three-dimensional model of the power equipment, to realize real-time reconstruction of the state of the power equipment, as follows: The three-dimensional model of the device is voxelized, and the model space is divided into × × uniform grids, 、 、 is a positive integer, and the center point coordinates of each voxel are generated ; The center point coordinates are: wherein, are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , are the maximum boundary points of the device three-dimensional model on the x, y, z axes respectively, x , y , z are the minimum boundary points of the device three-dimensional model on the x, y, z axes respectively, , , , are the intervals between the center points of two adjacent voxels on the x, y, z axes respectively; for each center point coordinate of a voxel, , the finite element hexahedron unit in which the center point is located is found, and the coordinates of the 8 vertices of the finite element hexahedron unit and the corresponding physical values and are read from the obtained result data file; ; The physical quantity values ​​of the eight vertices are weighted and summed, and the physical quantity values ​​of the voxel points are calculated using trilinear interpolation. physical quantity value for: wherein, are respectively voxel points are relative distances on x, y, z axes, is a weight value, the weight value The specific expression is as follows: ； After the obtained physical quantity values of each voxel point are normalized, the voxel points are mapped to a color space to generate a cloud chart of the voxel grid, to complete real-time reconstruction of the state of the power equipment.

17. A computer-readable storage medium storing one or more programs, the one or more programs comprising instructions for: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method of any one of claims 1-10.

18. A computing device, comprising: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method of any one of claims 1-10. One or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including programs for performing the steps of any of the methods of claims 1-10.

Images