Modular design method of ring main unit based on multi-cavity coupling

Abstract

The present invention discloses a modular design method for ring network cabinets based on multi-cavity coupling, which relates to the technical field of electrical equipment structural design, and includes the following steps: in the cavity structure simulation post-processing stage of finite element simulation software, the three-dimensional field variable data inside the cavity is extracted in real time through the software's built-in automated script interface, and the acquired data is exported in real time in a structured format to form a standardized data file. The present invention can accurately identify the risk of micro-eddy current circulation in the cavity through high-dimensional feature extraction, deep learning recognition and dynamic load response regulation. The structural eddy thermal anisotropy index and the electromagnetic dissipation focus index effectively quantify its focusing characteristics and thermoelectric anomalies, and the eddy current evolution path is analyzed in combination with load changes to construct an intelligent simulation system that can be verified in a closed loop, which helps to optimize the design of ring network cabinets for safety.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment structure design, and in particular to a modular design method for a ring network cabinet based on multi-cavity coupling. Background Art

[0002] "RMU modular design based on multi-cavity coupling" refers to the design of a RMU by dividing the switchgear interior into multiple functionally independent yet coupled structural cavities (such as busbar cavities, switch operating cavities, cable entry and exit cavities, and instrument control cavities). Adopting a modular design concept, each cavity is independently modeled, parameterized, and assembled as a standardized functional module. This approach typically utilizes 3D modeling software such as SolidWorks, Creo, and CATIA, combined with finite element simulation tools such as ANSYS and COMSOL, to conduct detailed simulation analysis of electromagnetic compatibility, thermal coupling effects, structural strength, and spatial layout between cavities. A modular assembly library manages each functional cavity within the design platform, allowing engineers to flexibly configure functional units based on diverse power system requirements and efficiently transition between customized design and standard batch manufacturing. Furthermore, module version control and collaborative management are implemented using PLM systems such as Teamcenter to ensure optimal overall design in terms of safety isolation, maintenance, and scalability.

[0003] Existing technologies have the following shortcomings: During actual operation, ring main units (RMUs) often operate under high-frequency dynamic electrical conditions. Especially during short-circuit shocks, arcing, or transient overvoltage events, tiny but continuous closed electromagnetic induction loops can form between multiple metal components within the cavity in local structural gaps, edge contact areas, or metal overlaps, generating micro-eddy currents. These areas are typically located at busbar junctions, switch housing connections, or between structural support metal parts. These areas are small, rapidly heat-accumulate, and are easily overlooked, making them typical "hidden junction areas."

[0004] However, when simulating electromagnetic thermal coupling between cavities using finite element analysis tools (such as ANSYS and COMSOL), if local mesh refinement is not used in the aforementioned microstructure areas, the simulation resolution will be insufficient to accurately capture the formation and evolution of these eddy current circulation paths. This can lead to a significant underestimation of heating, electric field concentration, or magnetic induction intensity in these areas in the calculation results, leading to potential risks such as underestimation of the thermal field and misjudgment of electrical breakdown risk. This problem is highly hidden and rarely occurs, but once it occurs, it can cause serious consequences such as insulation breakdown, localized corrosion, and even fire, seriously threatening the operational stability of the ring main unit and the safety of the system power supply.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a modular design method for ring main units (RMUs) based on multi-cavity coupling. Through high-dimensional feature extraction, deep learning recognition, and dynamic load response control, this method can accurately identify the risks of micro-eddy current circulation within the cavities. The structural eddy thermal anisotropy index and the electromagnetic dissipation focus index effectively quantify the focusing characteristics and thermoelectric anomalies. Combined with load variation, the eddy current evolution path is analyzed, and an intelligent simulation system with closed-loop verification is constructed to facilitate the safe optimization design of RMUs and address the above-mentioned background issues.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a modular design method for a ring main unit based on multi-cavity coupling, comprising the following steps:

[0008] During the post-processing phase of cavity structure simulation in the finite element simulation software, the 3D field variable data inside the cavity is extracted in real time through the built-in automated script interface of the software. The acquired data is then exported in real time in a structured format to form a standardized data file.

[0009] The obtained raw three-dimensional field variable data is preprocessed, and key characteristic indicators representing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data through feature engineering methods. The extracted key characteristic indicators are comprehensively analyzed to quantify the true probability and severity of the micro-eddy circulation phenomenon;

[0010] After formatting and data enhancement, the key characteristic indicators from the risk quantification analysis are input into a deep learning model that has been fully trained with historical simulation data and field operation data. The model automatically assesses the micro-eddy circulation risk of the cavity structure, achieving automated intelligent judgment of micro-eddy circulation risk.

[0011] When the potential risk of micro-eddy circulation in the cavity structure is identified, the load boundary conditions in the simulation environment are dynamically adjusted, and the dynamic evolution path of the eddy circulation is captured by performing a differential analysis of the transient evolution trends of the field variables under different load conditions.

[0012] Preferably, in the post-processing stage of the finite element simulation software, extracting the cavity structure three-dimensional field variable data and exporting it into a standardized file format includes the following steps:

[0013] Complete the modeling and simulation calculation of the cavity structure in the finite element simulation software to obtain the required three-dimensional field variable distribution results;

[0014] Call the software's built-in automated script interface and write a script to achieve batch extraction of target physical quantities in the entire cavity space;

[0015] The extracted three-dimensional field variable data is structured according to the correspondence between spatial coordinates and physical quantities to ensure that the data format is standardized and easy to process later;

[0016] Use scripts to export the sorted data into a standardized data file format.

[0017] Preferably, key characteristic indicators characterizing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data through feature engineering methods. The extracted key characteristic indicators include the degree of deviation of the directional change of temperature rise in unit volume and the concentration of electromagnetic energy consumption density in unit area in space. The degree of deviation of the directional change of temperature rise in unit volume and the concentration of electromagnetic energy consumption density in unit area in space are comprehensively analyzed under the monitoring window to generate structural eddy thermal anisotropy index and electromagnetic dissipation focus index respectively. The structural eddy thermal anisotropy index and electromagnetic dissipation focus index are used to quantify the true probability and severity of the micro-eddy circulation phenomenon.

[0018] Preferably, the specific steps of comprehensively analyzing the deviation degree of the directional change of the temperature rise within a unit volume in the monitoring window to generate the structural eddy thermal anisotropy index are as follows:

[0019] In the monitoring window, the unit volume unit is used as the basic analysis unit to extract the three-dimensional temperature rise field from the simulation data. The gradient components of the temperature rise along the three main directions are calculated respectively to form the temperature rise direction gradient vector field. The calculation formula of the temperature rise direction gradient vector field is as follows: , where is the gradient vector field in the direction of temperature rise, The temperature field is The spatial rate of change in the axial direction, The temperature field is The spatial rate of change in the axial direction, The temperature field is Spatial rate of change in the axial direction;

[0020] In each unit volume, the directional deviation tensor is calculated. The directional deviation tensor calculation expression is as follows: , where is the principal thermal conductivity direction vector of the structure, is the modulus of the principal thermal conductivity direction vector, is the modulus of the temperature rise gradient, is the directional deviation tensor value;

[0021] In order to amplify the structural differences caused by small deviations, Spatial coupling sensitivity processing is performed to highlight the contribution of the directional jump point to thermal anisotropy. The calculation expression of the structural eddy thermal anisotropy index is as follows: , where is the temperature Laplace modulus, is the temperature gradient modulus, It is an index of structural eddy thermal anisotropy.

[0022] Preferably, the specific steps of comprehensively analyzing the concentration of electromagnetic energy consumption density in a unit area in space under a monitoring window to generate an electromagnetic dissipation focus index are as follows:

[0023] In the finite element simulation environment, the three-dimensional electromagnetic energy consumption density distribution field of the unit area in the monitoring window is obtained through the simulation post-processing module. Based on the obtained electromagnetic energy consumption density distribution field, the local electromagnetic energy consumption focusing intensity index is calculated. The calculation expression is as follows: , where is the electromagnetic energy consumption density distribution field, is the space coordinate vector, is the gradient vector of the electromagnetic energy density, expressed at position The energy dissipation per unit volume at is the modulus of the gradient vector of the electromagnetic energy density, is the divergence operation of the vector field, It is an indicator of the local electromagnetic energy consumption focusing intensity;

[0024] Complete the local electromagnetic energy consumption focusing intensity index After calculation, select the focus intensity exceeding the preset threshold in the monitoring area The high-aggregation sub-region is defined as follows: ,in, It is a highly aggregated subregion. It is the three-dimensional space range covered by the monitoring window. is the focus intensity threshold;

[0025] Based on the obtained high-aggregation sub-regions, the electromagnetic dissipation focus index is calculated. The calculation expression is as follows: , where is the electromagnetic dissipation focus indicator, It is a volume differential element, representing a tiny volume unit in three-dimensional space.

[0026] Preferably, the structural eddy thermal anisotropy index and electromagnetic dissipation focus index after risk quantification analysis are formatted and data enhanced, and then input into a deep learning model that has been fully trained with historical simulation data and field operation data. The micro-eddy circulation risk coefficient is generated by the model, and the micro-eddy circulation risk of the cavity structure is automatically evaluated by the micro-eddy circulation risk coefficient, thereby realizing automated intelligent judgment of the micro-eddy circulation risk.

[0027] Preferably, the generated micro-eddy current circulation risk coefficient is compared and analyzed with a preset micro-eddy current circulation risk coefficient reference threshold value, and an intelligent judgment is made on whether there is a micro-eddy current circulation risk in the cavity structure. The judgment logic is as follows:

[0028] If the micro-eddy circulation risk coefficient is greater than the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is a potential micro-eddy circulation risk in the cavity structure; if the micro-eddy circulation risk coefficient is less than or equal to the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is no potential micro-eddy circulation risk in the cavity structure.

[0029] Preferably, after identifying the potential risk of micro-eddy circulation in the cavity structure, the load boundary conditions in the simulation environment are dynamically adjusted, and the transient evolution trends of the field variables under different load conditions are analyzed to capture the dynamic evolution path of the eddy circulation. The specific steps are as follows:

[0030] In the cavity area where potential micro-eddy current circulation risks have been identified, dynamic load boundary condition adjustment is implemented. By increasing the short-circuit current amplitude, improving the pulse waveform steepness, and extending the arc duration, three boundary disturbance methods are synergistically excited to form a composite working condition disturbance intensity index. The calculation formula is as follows: , where is the adjustment increment of the short-circuit current amplitude, is the pulse waveform steepness parameter, is the arc duration increment, is the reference short-circuit current amplitude, is the reference steepness of the pulse waveform, is the reference arc duration, 、 as well as are the adjustment increments of the short-circuit current amplitude Pulse waveform steepness parameter and the arc duration increment The weighted coefficient of , is the composite operating condition disturbance intensity index, is the micro-eddy circulation risk coefficient, is the reference threshold of the micro-eddy circulation risk coefficient, is the risk modulation sensitivity coefficient;

[0031] After implementing the dynamic load boundary condition adjustment, the eddy current circulation path is captured by the field variable dynamic evolution difference coefficient according to the field variable distribution changes under different load conditions. The calculation expression of the field variable dynamic evolution difference coefficient is as follows: , where is the spatial volume within the identified potential risk area, is the spatial gradient of magnetic flux density, which represents the magnetic flux density The rate of change in each direction in three-dimensional space, is the spatial gradient of the current density, representing the current density vector field The gradient changes in all directions in space, is the spatial gradient of temperature rise, is the magnetic flux density gradient weight coefficient, is the current density gradient weight coefficient, is the temperature rise gradient weight coefficient, is the volume differential element, is the coefficient of difference in the dynamic evolution of field variables;

[0032] Based on the obtained field variable dynamic evolution difference coefficient , an innovative spatial concentration index is constructed to accurately locate the degree of spatial aggregation of eddy circulation paths. The calculation formula is as follows: , where It is The spatial distance from the spatial subunit to the eddy response center, It is The difference coefficient of the dynamic evolution of the field variables of the spatial sub-units, is the total number of spatial subunits, is the spatial concentration index of the micro-eddy circulation path.

[0033] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0034] The present invention can achieve comprehensive identification and quantitative assessment of potential micro-eddy current circulation risks in cavity structures by introducing an analysis mechanism that combines high-dimensional field variable feature extraction, deep learning intelligent discrimination, and dynamic load response regulation. This method not only breaks through the problem of insufficient accuracy in identifying microstructural anomalies in traditional finite element simulation, but also accurately captures the focusing characteristics and thermoelectric response characteristics of eddy current circulation in space through innovative parameters such as structural eddy thermal anisotropy index and electromagnetic dissipation focus index. At the same time, with the help of dynamic adjustment and differential analysis of load boundary conditions, the induction mechanism and evolution path of eddy current risks are further revealed. Ultimately, a closed-loop technology system from data-driven identification, physical behavior modeling to risk verification feedback is constructed, providing a feasible and scalable intelligent simulation analysis path for the safe design, fault prediction and manufacturing optimization of ring network cabinet structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0036] Figure 1 This is a flow chart of the modular design method of a ring main unit based on multi-cavity coupling of the present invention. DETAILED DESCRIPTION

[0037] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0038] The present invention provides Figure 1 The modular design method of the ring main unit based on multi-cavity coupling shown includes the following steps:

[0039] During the post-processing phase of cavity structure simulation using finite element simulation software (such as ANSYS or COMSOL), the built-in automated scripting interface (e.g., APDL scripts in ANSYS, Java or Python API interfaces in COMSOL) is used to extract the 3D field variable data (e.g., magnetic flux density, current density, temperature rise distribution, etc.) inside the cavity in real time. The acquired data is then exported in a structured format in real time to form a standardized data file (e.g., CSV or HDF5 format files).

[0040] During the post-processing phase of finite element simulation, simulation results are programmatically manipulated using the automated programming interface (API) provided by the simulation software to automatically extract and export key physical field data. For example, in ANSYS, APDL (ANSYS Parametric Design Language) scripts can be used, while in COMSOL, simulation result modules can be called via Java or Python APIs to extract real-time 3D distribution data for physical quantities such as magnetic flux density, current density, temperature rise distribution, and electric field intensity. These scripting methods transform post-processing operations during the simulation process into automated tasks, significantly improving efficiency and data consistency.

[0041] Key physical field data from finite element simulation results is exported in a standardized and structured manner, forming files in common data formats such as CSV and HDF5, facilitating further processing and analysis in external data analysis platforms (such as Pandas and TensorFlow in Python). This data export not only extends simulation results beyond the software's internal graphical interface display but also transforms them into computable, trainable, and reusable data resources, laying a solid data foundation for subsequent risk identification, deep learning modeling, feature analysis, and dynamic control.

[0042] In the post-processing stage of the finite element simulation software, the three-dimensional field variable data of the cavity structure is extracted and exported into a standardized file format, including the following steps:

[0043] First, the cavity structure is modeled and simulated in finite element simulation software to obtain the required three-dimensional field variable distribution results.

[0044] Secondly, call the software's built-in automated script interface (such as ANSYS's APDL script or COMSOL's Java / Python API) to write a script to achieve batch extraction of target physical quantities (such as magnetic flux density, current density, temperature rise distribution, etc.) in the entire cavity space;

[0045] Third, the extracted three-dimensional field variable data is structured according to the correspondence between spatial coordinates and physical quantities to ensure that the data format is standardized and easy to process later.

[0046] Fourth, and finally, use scripts to export the collated data into standardized data file formats (such as CSV, HDF5) for further data analysis, modeling, or archiving on external platforms.

[0047] The obtained raw three-dimensional field variable data is preprocessed, and key characteristic indicators representing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data through feature engineering methods. The extracted key characteristic indicators are comprehensively analyzed to quantify the true probability and severity of the micro-eddy circulation phenomenon;

[0048] The field variable data obtained from finite element simulation often contain data noise and outliers due to numerical discretization, mesh errors, or low solver convergence accuracy. Therefore, before the data enters the analysis or modeling stage, systematic data preprocessing must be performed. This process mainly includes the following aspects:

[0049] The first step is data noise reduction and filtering. For spatially discrete data of three-dimensional physical field variables such as magnetic flux density, current density, and temperature rise distribution, a low-pass filter can be used to remove high-frequency spurious signal components while retaining the main trend information. Alternatively, a sliding window smoothing filter (such as moving average or weighted average) can be used to smooth local fluctuations, thereby reducing local glitches caused by factors such as dense discrete points and sudden gradient changes during the simulation process. For some abnormal sudden change points, an outlier removal method based on statistical thresholds (such as Z-Score or IQR discrimination) can be used to automatically detect and remove extreme values that do not conform to physical laws, thereby improving overall data quality.

[0050] The second important step is data normalization. Because different field variables (such as magnetic flux density in T, current density in A / m², and temperature rise in K) have vastly different dimensions and ranges, without normalization, subsequent feature extraction and deep learning model analysis will be unable to achieve a unified scale for comparison. Common methods include min-max scaling and z-score normalization, which map all data to the same interval or standard normal distribution, helping to balance the weights of different field variables.

[0051] Finally, format standardization is also a key step. This involves organizing multi-dimensional, multi-time, or multi-condition data into a unified structure (such as a tensor, DataFrame, or nested dictionary structure), and standardizing field naming, spatial coordinate mapping, and data order. This facilitates efficient call and batch processing during subsequent data fusion, feature engineering, or deep model training.

[0052] In general, the roles of the above preprocessing steps are: first, to improve the authenticity and stability of the original simulation data and remove interference signals; second, to enhance the comparability and integration between different physical variables, laying the foundation for building a multi-field coupling analysis model; third, to improve the convergence speed and recognition accuracy of subsequent AI model training and prediction, and to ensure the data reliability and engineering practicality of the entire simulation-identification system from the source.

[0053] Through feature engineering methods, key characteristic indicators characterizing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data. The extracted key characteristic indicators include the deviation degree of the directional change of temperature rise per unit volume and the spatial concentration degree of the electromagnetic energy consumption density per unit area. The deviation degree of the directional change of temperature rise per unit volume and the spatial concentration degree of the electromagnetic energy consumption density per unit area are comprehensively analyzed under the monitoring window to generate the structural eddy thermal anisotropy index and the electromagnetic dissipation focus index respectively. The structural eddy thermal anisotropy index and the electromagnetic dissipation focus index are used to quantify the true probability and severity of the micro-eddy circulation phenomenon.

[0054] The combined extraction and analysis of the structural eddy thermal anisotropy index and the electromagnetic dissipation focus index effectively quantifies the true probability and severity of micro-eddy circulation phenomena. This approach has a clear physical basis and is highly operational. The structural eddy thermal anisotropy index reflects the degree of uneven distribution of temperature rise gradients in different directions within a unit volume. If a region exhibits abnormally high deviations in the direction of heat conduction, it indicates that the heat source is likely not caused by a uniform load but rather by local electromagnetic induction heating (i.e., eddy currents), a characteristic of non-uniform, strongly directional heating. The electromagnetic dissipation focus index, on the other hand, measures the degree of focus of electromagnetic energy consumption within a unit area. When the electromagnetic energy consumption density is highly concentrated in space (rather than diffusely distributed), it generally indicates localized energy convergence, where eddy current energy accumulates in a confined area, leading to the risk of ablation or breakdown. Therefore, the structural eddy thermal anisotropy index can be used as a "spatial criterion for thermal anomalies," while the electromagnetic dissipation focus index serves as a "criterion for the intensity of electromagnetic dissipation focusing." Combined, the two can reflect the formation mechanism and destructive potential of eddy circulations from the perspectives of both thermal response mechanisms and electromagnetic driving forces. Through the spatiotemporal synchronous extraction and quantitative analysis of indicators under the monitoring window, it is possible to not only determine whether there is a vortex circulation risk, but also further estimate its evolution potential, damage level and the urgency of the response required, which has high engineering practicality and predictive value.

[0055] A significant deviation in the directional variation of temperature rise per unit volume can be a key indicator of the risk of micro-eddy current circulation within multiple metal components within a cavity. Under normal heat transfer conditions, the temperature rise is typically uniformly distributed along the main direction of thermal conductivity (such as along the axis of a metal plate or structure), and the thermal gradient exhibits clear continuity and isotropic consistency. However, when multiple conductive metal components within a cavity form closed eddy current paths due to electromagnetic induction coupling, the induced current circulates at high speed in localized regions, generating highly localized electromagnetic heating. This causes thermal energy to no longer diffuse uniformly, but instead concentrates in anomalous directions and rapidly heats up. This heating mechanism disrupts the directional consistency of heat conduction at the microscale, causing the temperature rise gradient in certain small regions to deviate significantly or jump, manifesting as anisotropic abrupt changes in the heat conduction field. Therefore, a significant deviation in the directional variation of temperature rise per unit volume often indicates the presence of a non-uniform energy input source in that region. This energy input is most likely caused by eddy current effects in the localized metal component coupling region, and serves as an important physical signal for determining the formation and evolution of micro-eddy current circulation.

[0056] The specific steps for comprehensively analyzing the deviation degree of the directional change of temperature rise within a unit volume under the monitoring window to generate the structural eddy thermal anisotropy index are as follows:

[0057] In the monitoring window, the unit volume unit is used as the basic analysis unit to extract the three-dimensional temperature rise field from the simulation data, and the temperature rise along the three main directions is calculated separately. The gradient component on the temperature rise direction forms the gradient vector field in the temperature rise direction. The calculation formula of the gradient vector field in the temperature rise direction is as follows: , where is the temperature rise direction gradient vector field, is the three-dimensional temperature field The first derivative in space represents the temperature rise along the three main directions The rate of change, is the temperature field exist The spatial rate of change in the axis direction, which indicates the unit distance along the The change in temperature when the direction moves, is the temperature field exist The spatial rate of change in the axis direction, which indicates the unit distance along the The change in temperature when the direction moves, is the temperature field exist The spatial rate of change in the axis direction, which indicates the unit distance along the The amount of temperature change when the direction moves;

[0058] In each unit volume, the directional deviation tensor is calculated to represent the degree of deviation between the temperature rise gradient vector at that point and the main structure thermal conductivity direction (such as the metal plate surface normal or cable routing direction). The directional deviation tensor calculation expression is as follows: , where It is the main heat conduction direction vector of the structure, representing the dominant heat conduction direction in the material or device under normal working conditions. It is used as a reference direction to compare the angle with the current temperature rise direction to determine whether the temperature rise is transmitted along the designed heat conduction channel. If the eddy current induces local temperature rise in the non-heat conduction path, and Will deviate significantly. is the modulus of the principal thermal conductivity direction vector, It is a unit vector with a constant modulus of 1. It is used to maintain the normalized ratio calculation. It can be omitted in practical applications (because it is always equal to 1). It is retained to fully express the normalized structure of the vector dot product. It is the modulus of the temperature rise gradient, the Euclidean norm of the temperature rise gradient vector in three-dimensional space, representing the intensity of heat flow change, independent of direction, and only reflects the spatial change speed of the temperature field at that point, and is used as a normalized benchmark. It is the directional deviation tensor value, which quantifies the degree of deviation between the temperature rise gradient direction within a unit volume and the main heat conduction direction of the structure. It is used to identify whether there is abnormal local heating caused by micro-eddy circulation. The value range is 0-2. The larger the value, the more the temperature rise direction at that location deviates from the main heat conduction direction of the structure. It can be used as an effective quantitative indicator of micro-eddy circulation risk.

[0059] The above steps extract the spatial variation characteristics of the local temperature rise conduction direction from the three-dimensional temperature rise field. By calculating the degree of deviation between the temperature rise gradient and the dominant heat direction of the structure, a quantitative index reflecting heat flow direction anomalies is constructed. This index can effectively identify unintended local heat flow deflection caused by micro-eddies within the cavity and is an important precursor to determining the potential paths of eddy current circulation.

[0060] In order to amplify the structural differences caused by small deviations, Spatial coupling sensitivity processing is performed to highlight the contribution of the directional jump point to thermal anisotropy. The calculation expression of the structural eddy thermal anisotropy index is as follows: , where It is the modulus of the temperature Laplace operator (i.e., the modulus of the second-order spatial derivative of temperature), which indicates the degree of curvature or mutation of the local temperature field in space. It is a key parameter to measure whether the temperature is rapidly gathering or dissipating at a certain point. It is used to capture the spatial mutation behavior in the temperature field. For example, eddy current heating causes the local temperature at a certain point to rise abnormally rapidly, or heat accumulates at structural defects. The larger the value, the more abnormal the temperature distribution at that point. It is the temperature gradient modulus (i.e., the first-order spatial derivative modulus of the temperature field), which indicates the speed of temperature change within a unit distance, that is, the instantaneous rate of change of temperature along the spatial direction, and is used for standardization. , to prevent misjudgment of local fluctuations due to large temperature base gradients. The "1 +" treatment in the formula is to avoid the denominator being zero while maintaining calculation stability. is the structural eddy thermal anisotropy index, The larger the value, the more serious the thermal conductivity discontinuity in the spatial direction of the region, which is most likely caused by micro-eddies.

[0061] By coupling the spatial second-order derivative (Laplacian operator) of the temperature field with the first-order gradient, this step amplifies the thermal discontinuity caused by sudden changes in curvature in localized regions due to deviations in the directionality of temperature rise, thereby highlighting the abnormal heat accumulation caused by eddy currents. This process enhances the ability to identify localized high energy density accumulation and thermal conductivity jumps caused by micro-eddy circulation, making the structural eddy thermal anisotropy index more sensitive and discernible in terms of spatial distribution.

[0062] The degree of deviation in the directional variation of temperature rise within a unit volume is comprehensively analyzed within the monitoring window to generate the structural eddy thermal anisotropy index. The structural eddy thermal anisotropy index quantifies the degree of deviation in the conduction of temperature rise within a unit volume in different spatial directions. A higher value indicates significant non-uniformity and strong directional deviation in the direction of temperature rise within that local area. Under normal operating conditions, heat conduction follows the isotropic or weakly anisotropic behavior of the material, and the directional deviation of temperature rise is minimal. However, when electromagnetic coupling generates micro-eddy currents within multiple metal components within the cavity, localized induction heating occurs along undesigned paths, causing heat to accumulate and diffuse along the induced current paths. This results in a sudden and asymmetric increase in the direction of the thermal gradient within the unit volume, significantly increasing the value of the structural eddy thermal anisotropy index. Therefore, a higher value of the structural eddy thermal anisotropy index indicates the presence of an abnormal energy input source, namely, micro-eddy currents. Conversely, when the value approaches the thermal field isotropic consistency benchmark, it indicates uniform heat conduction within the region and no signs of eddy current coupling within the structure.

[0063] High spatial concentrations of electromagnetic energy dissipation density per unit area typically indicate the risk of micro-eddy current circulation between multiple metal components within a cavity. This is because micro-eddy currents are essentially closed current paths formed by induced electromotive force within a conductor. Under high-frequency magnetic field perturbations, these induced currents form localized loops in gaps, overlapping edges, or uninsulated joints between components, causing energy to be repeatedly dissipated and converted into heat within a very confined space. When this electromagnetic energy dissipation density is no longer uniformly distributed in space but is highly concentrated in certain areas, with a high degree of overlap at the boundaries or junctions of metal structures, it generally indicates the presence of unintended induced current paths in these areas, a typical manifestation of eddy currents. Especially in asymmetric structural layouts or near coupling gaps, eddy current paths are more likely to be "confined" to a local area, resulting in an energy superposition effect that manifests as electromagnetic dissipation hotspots. Therefore, highly concentrated energy dissipation density not only reflects a significantly increased probability of eddy currents but also suggests the potential for engineering consequences such as structural overheating, insulation degradation, or material corrosion in this area. It is an important physical indicator for determining the risk of micro-eddy current circulation.

[0064] The specific steps for comprehensively analyzing the concentration of electromagnetic energy consumption density in space within a unit area under the monitoring window to generate the electromagnetic dissipation focus index are as follows:

[0065] In the finite element simulation environment, the three-dimensional electromagnetic energy consumption density distribution field of the unit area in the monitoring window is obtained through the simulation post-processing module. Based on the obtained electromagnetic energy consumption density distribution field, the local electromagnetic energy consumption focusing intensity index is calculated. The calculation expression is as follows: , where is the electromagnetic energy consumption density distribution field, expressed at a spatial point The electromagnetic energy consumption density at , that is, the energy loss per unit volume caused by electromagnetic effects (such as induced eddy currents, electric field coupling, etc.), is the space coordinate vector, is the gradient vector of the electromagnetic energy density, expressed at position The energy dissipation per unit volume at It is the modulus of the gradient vector of the electromagnetic energy consumption density, indicating the absolute strength of the gradient vector (i.e., the gradient size). If the value is larger, it means that the energy consumption changes dramatically near this point, and there may be a possible "energy section" or inflection point. It is the divergence operation of the vector field, which is used to calculate the "divergence" or "focusing" ability of the vector field at a certain point in space. If the divergence value is positive, it indicates that the energy is "diffused outward"; if it is negative, it indicates that the energy is "gathered inward". This formula takes its absolute value to capture the magnitude of the focusing intensity without distinguishing the direction. It is an indicator of the local electromagnetic energy consumption focusing intensity, indicating the spatial position The concentration intensity of electromagnetic energy consumption density at a location is used to measure whether electromagnetic energy consumption is spatially focused at that location, and can also be understood as "whether the local energy density is concentrated at that point";

[0066] The above steps are to couple the energy consumption density with its directional normalized gradient and then take the divergence to characterize the aggregation trend of local energy density. The larger the value, the stronger the local concentration of energy distribution near the spatial point, which is a sensitive characteristic indicator of the potential eddy current coupling focus.

[0067] Complete the local electromagnetic energy consumption focusing intensity index After calculation, select the focus intensity exceeding the preset threshold in the monitoring area The high-aggregation sub-region is defined as follows: ,in, is a high-aggregation sub-region, representing the entire monitoring area In the local electromagnetic energy consumption focusing intensity index Exceeding the preset threshold The area set composed of all points is used to extract the location area with the strongest energy focus and the most likely eddy current risk. It is the three-dimensional space range covered by the monitoring window. is the focus intensity threshold, which is used to divide the critical value of "high concentration area" and "normal area";

[0068] Based on the obtained high-aggregation sub-regions, the electromagnetic dissipation focus index is calculated. The calculation expression is as follows: , where is the electromagnetic dissipation focus indicator, It is a volume differential element, representing a tiny volume unit in three-dimensional space. It is used to sum spatial continuous variables in integral calculations and to perform volume-weighted accumulation of the focusing intensity or energy consumption density of the entire area.

[0069] In the above formula, the numerator represents the total amount of energy focused in the area with significant energy concentration intensity, and the denominator represents the total amount of electromagnetic energy consumed in the entire area. The obtained ratio is This is the electromagnetic dissipation focus index described in the present invention. Its physical meaning is: when The larger the value, the higher the proportion of unit energy consumption is concentrated in the local structural area, which is very likely to induce micro-eddy circulation phenomenon; on the contrary, The smaller the value, the more uniform the energy consumption distribution is and the lower the eddy current risk is.

[0070] The concentration of electromagnetic energy consumption density in space within a unit area is comprehensively analyzed under the monitoring window to generate an electromagnetic dissipation focus index. The electromagnetic dissipation focus index reflects the degree of concentration of electromagnetic energy consumption density in space within a unit area, that is, whether the energy is highly concentrated in a local area rather than evenly distributed. The larger the performance value of this index, the more obvious the accumulation of electromagnetic energy in the local area. This is usually due to the formation of undesigned closed induction paths between multiple metal components - that is, micro-eddy current circulation - which prevents electromagnetic energy from effectively diffusing in space. Instead, it is repeatedly dissipated and converted into heat in a small area. On the contrary, if the electromagnetic dissipation focus index is small, it means that the energy consumption is more evenly distributed in the structure, and there are no abnormally coupled eddy current paths between the metal components, indicating that the structure is in a normal electromagnetic response state and there is no risk of micro-eddy current circulation.

[0071] After formatting and data enhancement, the key characteristic indicators from the risk quantification analysis are input into a deep learning model that has been fully trained with historical simulation data and field operation data. The model automatically assesses the micro-eddy circulation risk of the cavity structure, achieving automated intelligent judgment of micro-eddy circulation risk.

[0072] After format adjustment and data enhancement processing, the structural eddy thermal anisotropy index and electromagnetic dissipation focus index after risk quantification analysis are input into a deep learning model that has been fully trained with historical simulation data and field operation data. The micro-eddy circulation risk coefficient is generated by the model, and the micro-eddy circulation risk of the cavity structure is automatically evaluated by the micro-eddy circulation risk coefficient, thereby realizing automated intelligent judgment of the micro-eddy circulation risk.

[0073] Deep learning models, pre-trained with historical simulation data and field operation data, are AI models (e.g., convolutional neural networks (CNNs), long short-term memory (LSTMs), or Transformer architectures) built and optimized based on extensive historical finite element simulation datasets and actual equipment operation monitoring data before formally deploying them for micro-eddy current risk assessment. These models are capable of identifying the mapping relationship between complex field variable patterns and risk characteristics. During the training phase, developers first input a large amount of labeled data (i.e., simulation samples and actual fault instances with confirmed micro-eddy currents). Through feature extraction, data normalization, and enhancement, the model is fed with high-dimensional feature vectors such as the structural eddy thermal anisotropy index (AHT-Index) and the electromagnetic dissipative focus index (EDF-Index). Through iterative optimization and backpropagation, the model continuously adjusts network weights to learn how to effectively match these spatial field variable characteristics with actual risk conditions, ultimately enabling the model to accurately output risk assessment results for new, unknown samples.

[0074] "Fully trained" in this process means the model has learned sufficient characteristic distribution patterns and risk evolution trends from data samples covering diverse boundary conditions, including various structural layouts, current excitations, material parameters, and external disturbance variations. Furthermore, online monitoring data from industrial sites (such as infrared thermal imaging, local temperature rise sensor data, and current mutation records) may be integrated to enhance the model's robustness and generalization to non-ideal disturbances under real-world operating conditions. After sufficient training, the deep learning model can perform rapid, automated, and probabilistic risk assessments on new input data (i.e., new simulation samples). It outputs a quantifiable "micro-eddy current circulation risk coefficient" (which can be a probability value between 0 and 1 or a multi-level risk score). Based on this coefficient, it determines whether high-risk circulation hazards exist within the cavity, the severity of the risk, and whether subsequent structural optimization or operational warnings are necessary. This intelligent assessment method effectively replaces traditional methods that rely on manual judgment and offers high efficiency, real-time performance, and mass scalability. It represents a key technological approach for risk control in modern electrical structure simulations.

[0075] The generated micro-eddy current risk coefficient is compared with the pre-set micro-eddy current risk coefficient reference threshold value to make an intelligent judgment on whether there is a micro-eddy current risk in the cavity structure. The judgment logic is as follows:

[0076] If the micro-eddy circulation risk coefficient is greater than the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is a potential micro-eddy circulation risk in the cavity structure; if the micro-eddy circulation risk coefficient is less than or equal to the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is no potential micro-eddy circulation risk in the cavity structure.

[0077] When potential micro-eddy current circulation risks are identified in the cavity structure, the load boundary conditions (such as short-circuit current amplitude, pulse waveform, and arc duration) in the simulation environment are dynamically adjusted. By performing differential analysis on the transient evolution trends of field variables (such as magnetic flux density, current density, and temperature rise distribution) under different load conditions, the dynamic evolution path of the eddy current circulation is captured;

[0078] By dynamically adjusting the load boundary conditions in the simulation environment, the potential micro-eddy circulation response behavior is actively stimulated, thereby enhancing its observability and difference characteristics in the simulation results, so as to achieve accurate identification and traceability analysis of the dynamic evolution path of the eddy circulation.

[0079] After initially identifying areas in the cavity structure where there is a risk of eddy current circulation, the distribution of field variables under a single static simulation condition can often only provide limited information and cannot fully reveal the complete dynamic process of eddy currents from "germination" to "expansion". Especially inside cavities with complex structures and staggered arrangements of multiple metal components, the generation mechanism of micro eddy current circulation usually relies on high-frequency electromagnetic excitation or non-steady-state disturbance environment. Therefore, introducing a multi-condition dynamic excitation mechanism in the simulation, such as increasing the short-circuit current amplitude, changing the rising edge steepness of the arc pulse waveform, or extending its duration, can artificially amplify the induced current density, magnetic flux disturbance, and energy coupling effects, so that the eddy current path that was originally in a weak coupling or boundary state shows a significant response in the electromagnetic and thermal fields.

[0080] By performing time-series sampling and differential analysis on key field variables such as magnetic flux density, current density, and temperature rise distribution under different load conditions, it is possible to dynamically capture the evolution trend and mutation behavior of these variables in high-risk areas, and further deduce the formation path, expansion direction, and focal area of the eddy current circulation. In particular, when comparing the response differences of multiple excitation conditions, the nonlinear enhancement effect of the eddy current response can be revealed, providing a basis for judging its development severity and critical transition. In addition, dynamic simulation can also assist in verifying the authenticity of the risk area and prevent misjudgment due to initial simulation modeling errors, thereby providing strong support in structural optimization, fault prediction, and simulation modeling accuracy control. In summary, this step is the key link in transforming "passive identification" into "active excitation + dynamic tracking", and is the core technical support for achieving accurate quantification of micro-eddy current risk mechanisms and path analysis.

[0081] After identifying the potential risk of micro-eddy circulation in the cavity structure, the load boundary conditions in the simulation environment are dynamically adjusted. By performing a differential analysis of the transient evolution trends of field variables under different load conditions, the specific steps to capture the dynamic evolution path of the eddy circulation are as follows:

[0082] In the cavity area where potential micro-eddy current circulation risks have been identified, dynamic load boundary condition adjustment is implemented. By increasing the short-circuit current amplitude, improving the pulse waveform steepness, and extending the arc duration, three boundary disturbance methods are synergistically excited to form a composite working condition disturbance intensity index. The calculation formula is as follows: , where is the adjustment increment of the short-circuit current amplitude, which indicates how much the short-circuit current under the current excitation increases compared with the original simulation conditions. It is the pulse waveform steepness parameter, which indicates the speed of the rising edge of the current or voltage excitation signal. The larger the value, the more rapid the signal change. is the arc duration increment, which indicates how much longer the arc action time applied in this simulation is than that in the reference case. It is the reference short-circuit current amplitude, which is the benchmark current value. It usually comes from the standard simulation model, industry specified value or rated short-circuit current under normal working conditions. , is the reference steepness of the pulse waveform, which represents the steepness parameter under the default pulse condition, is the reference arc duration, which is the set value of the arc duration under the standard simulation model. 、 as well as are the adjustment increments of the short-circuit current amplitude Pulse waveform steepness parameter and the arc duration increment The weighted coefficient of , It is the disturbance intensity index of composite working conditions. It is a quantitative indicator of the degree of enhanced excitation of simulation boundary conditions in finite element simulation by identifying its evolution path. It is used to characterize the "disturbance injection intensity" of simulation excitation on system response and introduces risk coefficient modulation to form a quantifiable and controllable dynamic excitation intensity control variable. is the micro-eddy circulation risk coefficient, is the reference threshold of the micro-eddy circulation risk coefficient, It is the risk modulation sensitivity coefficient, which controls the amplification factor of the risk coefficient deviation on the overall incentive intensity. The larger the value, the more sensitive it is to risk changes.

[0083] The purpose of this step is to actively and significantly amplify the eddy current response phenomenon by precisely adjusting the load conditions in the simulation environment, so as to highlight the different changes in the eddy current circulation path under different working conditions. It will be used as the measure of the excitation condition in the next step of analyzing the eddy current response.

[0084] After implementing the dynamic load boundary condition adjustment, the eddy current circulation path is captured by the field variable dynamic evolution difference coefficient according to the field variable distribution changes under different load conditions. The calculation expression of the field variable dynamic evolution difference coefficient is as follows: , where It is the spatial volume within the identified potential risk area. It refers to the local spatial area with potential micro-eddy current circulation risk identified by structural risk screening, thermal / electrical / magnetic feature focusing, etc. during the simulation post-processing process. It is usually a three-dimensional sub-area at the junction of multiple metal components or the narrow neck of the structure. is the spatial gradient of magnetic flux density, which represents the magnetic flux density The rate of change in each direction in three-dimensional space is used to reflect the degree of spatial fluctuation of the local magnetic field and is a key indicator for evaluating whether induced current may be induced. is the spatial gradient of the current density, representing the current density vector field The gradient changes in various directions in space reflect the concentration or diffusion state of the induced current inside the conductor. It is used to determine whether the induced current is enhanced along a closed path and whether it changes dramatically in a specific area. It is the spatial gradient of temperature rise, which represents the gradient of the temperature field distribution change in three-dimensional space, especially the spatial derivative of the temperature rise mutation area, which is used to judge whether the heat in the cavity shows abnormal non-uniform diffusion. is the magnetic flux density gradient weight coefficient, which indicates the weight distribution of the magnetic field gradient in the total response. For high-frequency excitation conditions, it should be appropriately increased. , such as setting it to 0.5-0.6, is the current density gradient weight coefficient, which represents the weight of the contribution of the current density spatial variation to the eddy current path formation. If the simulation goal is to identify the current induced circulation path, it is more appropriate to set it to 0.2-0.4. is the temperature rise gradient weight coefficient, which indicates the influence of temperature change gradient on the comprehensive response. For sealed gas insulated cabins, Set to 0.1-0.3 to emphasize temperature sensitivity and meet , is the volume differential element, It is the coefficient of difference in the dynamic evolution of field variables. It is a composite indicator used to quantitatively analyze the degree of difference in the spatial variation trends of field variables such as magnetic flux density, current density, and temperature rise under different load conditions. The larger the value, the more drastic the changes in magnetic field, current, and temperature in this area when excited.

[0085] The purpose of this step is to comprehensively consider the spatial gradient distribution characteristics of the three key field variables (magnetic field, current, and temperature) and quantitatively capture the evolution trend of the field variables in the micro-eddy circulation area under dynamic working conditions. The significant increase in the value reveals the dynamic formation and development characteristics of the eddy circulation path. The specific value of this coefficient will be further used to identify the core position of the circulation path in the next step.

[0086] Based on the obtained field variable dynamic evolution difference coefficient , an innovative spatial concentration index is constructed to accurately locate the degree of spatial aggregation of eddy circulation paths. The calculation formula is as follows: , where It is The spatial distance from the spatial sub-unit to the eddy response center refers to the current unit The Euclidean distance relative to the center of the eddy current response characterizes the degree to which the point deviates from the center of the eddy current response in space and is used to determine whether the response is highly focused. It is The difference coefficient of the dynamic evolution of the field variables of the spatial sub-units, is the total number of spatial subunits, It is the spatial concentration index of the micro-eddy circulation path, which indicates the "distribution radius" or "diffusion center" of the eddy circulation response in space. It is used to quantitatively evaluate whether it is concentrated. The smaller the value, the more concentrated the micro-eddy circulation response is in a small area, which is manifested as eddy energy focusing and local enhancement. This situation is more risky and can easily cause serious problems such as local overheating, breakdown or structural corrosion. The larger the value, the more dispersed the eddy response is in a larger spatial range, the path is in a "diffuse" state, and the energy density is lower than the concentrated state. It is not easy to form destructive consequences at a certain location, but it may cause system-level energy loss or latent heat accumulation.

[0087] The purpose of this step is to combine the dynamic evolution trend of eddy circulation with spatial aggregation, and calculate the spatial concentration index , which determines the spatial concentration of vortex circulation paths. Smaller index values indicate a more concentrated vortex circulation path in a specific area; larger index values indicate a more widespread path. This index can further clarify the specific location and severity of micro-vortex circulations, providing effective support for subsequent structural optimization and safety assurance.

[0088] The present invention can achieve comprehensive identification and quantitative assessment of potential micro-eddy current circulation risks in cavity structures by introducing an analysis mechanism that combines high-dimensional field variable feature extraction, deep learning intelligent discrimination, and dynamic load response regulation. This method not only breaks through the problem of insufficient accuracy in identifying microstructural anomalies in traditional finite element simulation, but also accurately captures the focusing characteristics and thermoelectric response characteristics of eddy current circulation in space through innovative parameters such as structural eddy thermal anisotropy index and electromagnetic dissipation focus index. At the same time, with the help of dynamic adjustment and differential analysis of load boundary conditions, the induction mechanism and evolution path of eddy current risks are further revealed. Ultimately, a closed-loop technology system from data-driven identification, physical behavior modeling to risk verification feedback is constructed, providing a feasible and scalable intelligent simulation analysis path for the safe design, fault prediction and manufacturing optimization of ring network cabinet structures.

[0089] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0090] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0091] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0092] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0093] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0096] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0097] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0098] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. A modular design method for ring main units based on multi-cavity coupling, characterized in that: The following steps are involved: During the post-processing phase of cavity structure simulation in the finite element simulation software, the 3D field variable data inside the cavity is extracted in real time through the built-in automated script interface of the software. The acquired data is then exported in real time in a structured format to form a standardized data file. The obtained raw three-dimensional field variable data is preprocessed, and key characteristic indicators representing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data through feature engineering methods. The extracted key characteristic indicators are comprehensively analyzed to quantify the true probability and severity of the micro-eddy circulation phenomenon; After formatting and data enhancement, the key characteristic indicators from the risk quantification analysis are input into a deep learning model that has been fully trained with historical simulation data and field operation data. The model automatically assesses the micro-eddy circulation risk of the cavity structure, achieving automated intelligent judgment of micro-eddy circulation risk. When potential micro-eddy circulation risks are identified in the cavity structure, the load boundary conditions in the simulation environment are dynamically adjusted. By performing differential analysis on the transient evolution trends of field variables under different load conditions, the dynamic evolution path of the eddy circulation is captured; Through feature engineering methods, key characteristic indicators characterizing the potential risk of micro-eddy circulation are extracted from the preprocessed high-dimensional field variable data. The extracted key characteristic indicators include the deviation degree of the directional change of temperature rise per unit volume and the spatial concentration degree of the electromagnetic energy consumption density per unit area. The deviation degree of the directional change of temperature rise per unit volume and the spatial concentration degree of the electromagnetic energy consumption density per unit area are comprehensively analyzed under the monitoring window to generate the structural eddy thermal anisotropy index and the electromagnetic dissipation focus index respectively. The structural eddy thermal anisotropy index and the electromagnetic dissipation focus index are used to quantify the true probability and severity of the micro-eddy circulation phenomenon.

2. The modular design method of ring main unit based on multi-cavity coupling according to claim 1 is characterized in that: In the post-processing stage of the finite element simulation software, the three-dimensional field variable data of the cavity structure is extracted and exported into a standardized file format, including the following steps: Complete the modeling and simulation calculation of the cavity structure in the finite element simulation software to obtain the required three-dimensional field variable distribution results; Call the software's built-in automated script interface and write a script to achieve batch extraction of target physical quantities in the entire cavity space; The extracted three-dimensional field variable data is structured according to the correspondence between spatial coordinates and physical quantities to ensure that the data format is standardized and easy to process later; Use scripts to export the sorted data into a standardized data file format.

3. The modular design method of ring main unit based on multi-cavity coupling according to claim 1 is characterized in that: The specific steps for comprehensively analyzing the deviation degree of the directional change of temperature rise within a unit volume under the monitoring window to generate the structural eddy thermal anisotropy index are as follows: In the monitoring window, the unit volume unit is used as the basic analysis unit to extract the three-dimensional temperature rise field from the simulation data. The gradient components of the temperature rise along the three main directions are calculated respectively to form the temperature rise direction gradient vector field. The calculation formula of the temperature rise direction gradient vector field is as follows: , where is the gradient vector field in the direction of temperature rise, The temperature field is The spatial rate of change in the axial direction, The temperature field is The spatial rate of change in the axial direction, The temperature field is Spatial rate of change in the axial direction; In each unit volume, the directional deviation tensor is calculated. The directional deviation tensor calculation expression is as follows: , where is the principal thermal conductivity direction vector of the structure, is the modulus of the principal thermal conductivity direction vector, is the modulus of the temperature rise gradient, is the directional deviation tensor value; In order to amplify the structural differences caused by small deviations, Spatial coupling sensitivity processing is performed to highlight the contribution of the directional jump point to thermal anisotropy. The calculation expression of the structural eddy thermal anisotropy index is as follows: , where is the temperature Laplace modulus, is the temperature gradient modulus, It is an index of structural eddy thermal anisotropy.

4. The modular design method of ring main unit based on multi-cavity coupling according to claim 1 is characterized in that: The specific steps for comprehensively analyzing the concentration of electromagnetic energy consumption density in space within a unit area under the monitoring window to generate the electromagnetic dissipation focus index are as follows: In the finite element simulation environment, the three-dimensional electromagnetic energy consumption density distribution field of the unit area in the monitoring window is obtained through the simulation post-processing module. Based on the obtained electromagnetic energy consumption density distribution field, the local electromagnetic energy consumption focusing intensity index is calculated. The calculation expression is as follows: , where is the electromagnetic energy consumption density distribution field, is the space coordinate vector, is the gradient vector of the electromagnetic energy density, expressed at position The energy dissipation per unit volume at is the modulus of the gradient vector of the electromagnetic energy density, is the divergence operation of the vector field, It is an indicator of the local electromagnetic energy consumption focusing intensity; Complete the local electromagnetic energy consumption focusing intensity index After calculation, select the focus intensity exceeding the preset threshold in the monitoring area The high-aggregation sub-region is defined as follows: ,in, It is a highly aggregated subregion. It is the three-dimensional space range covered by the monitoring window. is the focus intensity threshold; Based on the obtained high-aggregation sub-regions, the electromagnetic dissipation focus index is calculated. The calculation expression is as follows: , where is the electromagnetic dissipation focus indicator, It is a volume differential element, representing a tiny volume unit in three-dimensional space.

5. The modular design method of ring main unit based on multi-cavity coupling according to claim 1 is characterized in that: After format adjustment and data enhancement processing, the structural eddy thermal anisotropy index and electromagnetic dissipation focus index after risk quantification analysis are input into a deep learning model that has been fully trained with historical simulation data and field operation data. The micro-eddy circulation risk coefficient is generated by the model, and the micro-eddy circulation risk of the cavity structure is automatically evaluated by the micro-eddy circulation risk coefficient, thereby realizing automated intelligent judgment of the micro-eddy circulation risk.

6. The modular design method for ring main unit based on multi-cavity coupling according to claim 5 is characterized in that: The generated micro-eddy current risk coefficient is compared with the pre-set micro-eddy current risk coefficient reference threshold value to make an intelligent judgment on whether there is a micro-eddy current risk in the cavity structure. The judgment logic is as follows: If the micro-eddy circulation risk coefficient is greater than the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is a potential micro-eddy circulation risk in the cavity structure; if the micro-eddy circulation risk coefficient is less than or equal to the pre-set micro-eddy circulation risk coefficient reference threshold, it is judged that there is no potential micro-eddy circulation risk in the cavity structure.

7. The modular design method for ring main unit based on multi-cavity coupling according to claim 6 is characterized in that: After identifying the potential risk of micro-eddy circulation in the cavity structure, the load boundary conditions in the simulation environment are dynamically adjusted. By performing a differential analysis of the transient evolution trends of field variables under different load conditions, the specific steps to capture the dynamic evolution path of the eddy circulation are as follows: In the cavity area where potential micro-eddy current circulation risks have been identified, dynamic load boundary condition adjustment is implemented. By increasing the short-circuit current amplitude, improving the pulse waveform steepness, and extending the arc duration, three boundary disturbance methods are synergistically excited to form a composite working condition disturbance intensity index. The calculation formula is as follows: , where is the adjustment increment of the short-circuit current amplitude, is the pulse waveform steepness parameter, is the arc duration increment, is the reference short-circuit current amplitude, is the reference steepness of the pulse waveform, is the reference arc duration, 、 as well as are the adjustment increments of the short-circuit current amplitude Pulse waveform steepness parameter and the arc duration increment The weighted coefficient of , is the composite operating condition disturbance intensity index, is the micro-eddy circulation risk coefficient, is the reference threshold of the micro-eddy circulation risk coefficient, is the risk modulation sensitivity coefficient; After implementing the dynamic load boundary condition adjustment, the eddy current circulation path is captured by the field variable dynamic evolution difference coefficient according to the field variable distribution changes under different load conditions. The calculation expression of the field variable dynamic evolution difference coefficient is as follows: , where is the spatial volume within the identified potential risk area, is the spatial gradient of magnetic flux density, which represents the magnetic flux density The rate of change in each direction in three-dimensional space, is the spatial gradient of the current density, representing the current density vector field The gradient changes in all directions in space, is the spatial gradient of temperature rise, is the magnetic flux density gradient weight coefficient, is the current density gradient weight coefficient, is the temperature rise gradient weight coefficient, is the volume differential element, is the coefficient of difference in the dynamic evolution of field variables; Based on the obtained field variable dynamic evolution difference coefficient , an innovative spatial concentration index is constructed to accurately locate the degree of spatial aggregation of eddy circulation paths. The calculation formula is as follows: , where It is The spatial distance from the spatial subunit to the eddy response center, It is The difference coefficient of the dynamic evolution of the field variables of the spatial sub-units, is the total number of spatial subunits, is the spatial concentration index of the micro-eddy circulation path.

Images